Household Energy Handbook

- -- -

WORLD BANK TECHNICAL PAPER NUMBER 67

Household Energy Handbook An Interim Guide and Reference Manual

Gerald Leach and Marcia Gowen

under the guidance of Richard Dosik Rene Moreno

Willem Floor Mikael Grut Fernando Manibog and Kenneth Newcombe

The World Bank Washington DC

The International Bank for Reconstruction and DevelopmentTHE WORLD BANK

1818 H Street NW Washington DC 20433 USA

Technical Papers are not fonnal publications of the World Bank and are circulated to encourage discussion and comment and to communicate the results of the Banks work qUickly to the development community citation and the use of these papers should take account of their provisional character The findings interpretations and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank to its affiliated organizations or to members of its Board of Executive Directors or the countries they represent Any maps that accompany the text have been prepared solely for the convenience of readers the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank its affiliates or its Board or member countries concerning the legal status of any country territory city or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation

Because of the infonnality and to present the results of research with the least possible delay the typescript has not been prepared in accordance with the procedures appropriate to formal printed text-s and the World Bank accepts no responsibility for errors The publication is supplied at a token charge to defray part of the cost of manufacture and distribution

The most recent World Bank publications are described in the catalog New Publications a new edition of which is issued in the spring and fall of each year The complete backlist of publications is shown in the annual Index of Publications which contains an alphabetical title list and indexes of subjects authors and countries and regions it is of value principally to libraries and institutional purchasers The latest edition of each of these is available free of charge from the Publications Sales Unit Department F The World Bank 1818 H Street NW Washington DC 20433 USA or from Publications The World Bank 66 avenue dUna 75116 Paris France

Gerald Leach is senior fellow at the International Institute for Environment and Development London Marcia Gowen is a fellow at the Resource Systems Institute of the East-West Center Honolulu

Library of Congress Cataloging-in-Publication Data Leach Gerald

Household energy handbook

(World Bank technical paper ISSN 0253-7494 no 67) Bibliography p 1 Dwe11ings--Deve1oping countries--Energy

conservation 2 Power resources--Deve1oping countries I Gowen Marcia M 1954shyII Title III Series TJ1635D86L43 1987 33379 87-18864 ISBN 0-8213-0937-4

- iii shy

ABSTRACT

Traditional household fuels play a vital role in developing countries More than two billion people depend on them to meet basic energy needs Today many of these people are facing a deepening crisis of energy scarcity as local wood resources are depleted and more distant forests are cut down The implications of this crisis extend beyond the supply of energy itself As trees are lost the land which provides their livelihood and feeds the nation may become more vulnerable to erosion and soil degradation In some arid parts of the developing world this process has reached the terminal stage where the land produces nothing and starvation or migration are the only alternatives

Much needs to be done to address the household energy problems of the developing countries Household energy use must be made more efficient Fuel substitution must be encouraged Wood and other energy supplies must be augmented and priced affordably However to successfully implement these remedies requires a sound understanding of the basic supply and demand variables operating in the sector These variables have been difficult to measure because traditional fuels are frequently not traded and because of the large variation in the availability and costs of energy supplies in the levels and trends of consumption and mix of fuels employed in end-uses technologies and energy-related preferences and modes of behavior

A standard framework for measuring and assessing technical information on the household energy sector is needed to more adequately address these difficulties This handbook is intended as a first step toward creating such a framework Chapter I discusses energy terms and principles underlying the energy units definitions and calculations presented in the following chapters Chapter II describes household consumption patterns and their relationship to income location and household-size variables Chapter III evaluates energy end-uses and the technologies which provide cooking lighting refrigeration and space heating services Chater IV examines household energy resources and supplies focusing on traditional biomass fuels Finally Chapter V demonstrates simple assessment methods and presents case studies to illustrate how household energy data can be used in different types of assessments

- iv shy

This report is based primarily on the work of its principal authors Gerald Leach and Marcia Gowen From inception to completion of the report the authors received guidance from a Review Committee consisting of Richard Dosik Rene Moreno WiUem Floor Mikael Grut Fernando Manibog and Kenneth Newcombe who made many contributions The report also benefited from the valuable comments received from experts outside the World Bank Russell deLucia (deLucia and Associates) MR de Montalembert (F AO) and Krishna Prasad (Eindhoven University of Technology) Collectively staff in the World Bank Energy Department contributed significantly with comments and suggestions at various stages in the production of the Handbook Matthew Mendis Dale Gray and Robert van der Plas deserve particular mention The final manuscript was greatly enhanced by the expert creative editing of Maryellen Buchanan Linda Walker-Adigwe provided outstanding word processing support

- v shy

TABLE or CORTEIITS

INTRODUCTION 1 The Importance of Household Energy in Developing countries 1 Characteristics of Household Energy 2 Purpose of the Handbook 4 Organization of the Handbook 4

CHAPTER I ENERGY MEASUREMENT AND DEFINITIONSbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 6

CHAPTER II HOUSEHOLD ENERGY CONSUMPTION bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 28

A Objectives and Structure bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 6 B Basic Measurement Concepts bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 6

Measurement Systems and Reference Data bullbullbullbullbullbull 6 Production and Conversion Systems bullbullbullbullbullbullbullbullbullbullbull 6 Measurement Units bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 9 Gross and Net Heating Values bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 9 Heating Values and Moisture Content bullbullbullbullbullbullbullbullbull 11 Volume Density and Moisture Content bullbullbullbullbullbullbullbull 16

C Utilized Energy Efficiency and Specific Fuel Consumptionbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 19

Primary and Delivered Energy Efficiencies bullbullbull 19 Definitions of Efficiencybullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 20 Specific Fuel Consumption Energy

Intensity and Fuel Economybullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 22 D Basic Statistics bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 24

Data Validitybullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 24 Elasticities bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 25

A Objectives and Structure bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 28 B Data Resources bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 29

National Energy Balances bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 29 National Budget Surveys bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 29 National Energy Surveys bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 31 Local Micro Surveys bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 31

C Major Consumption Variables bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 33 Gathered Fuels and Time Budgets bullbullbullbullbullbullbullbullbullbullbullbullbull 37 Time Costs of Fuel Collectionbullbullbullbullbullbullbullbullbullbullbullbullbull 40 Income and Rural-Urban Differencesbullbullbullbullbullbullbullbullbullbull 41 Household Size bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 45 Purchased Fuels and Expenditure Shares bullbullbullbullbullbull SO Energy Prices bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 51

D Adaptations to Fuel Scarcitybullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 52 Adaptations in Rural Areas bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 52 Adaptations in Urban Areas bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 55

E Energy End-Uses bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull ~ bullbullbullbullbullbullbullbull 57 F Summarybullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 60

-~~~~ ~~---------------------~-------

vi

CHAPTER III

CHAPTER IV

ENERGY END-USES AND TECHNOLOGIES bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull A Objectives and Structurebullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull B Cooking bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

Consumption Ranges bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Fuel Preferences bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

C Cooking Stoves and Equipment bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Stove Types bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Stove Efficiencies and Fuel Savings bullbullbullbullbullbullbullbullbull Other Technical Aspects bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Stove Costs bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Dissemination and Impact bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

D Lighting bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Measurement Units and Standards bullbullbullbullbullbullbullbullbullbullbullbullbull Lighting Energy Fuels and Technologies bullbullbullbull Photovoltaic Lighting bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

E Refrigeration and Other Electrical End-Uses bullbullbull F Space Heating bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

HOUSEHOLD ENERGY SUPPLIES bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull A Objectives and Structurebullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull B Background Perspectives bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

Village Biomass Systems bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Access to Resources bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Involving the People bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Tree Loss and Tree Growingbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

C Fuelwood Resources and Productionbullbullbullbullbullbullbullbullbullbullbullbullbull Measurement Units and Concepts bullbullbullbullbullbullbullbullbullbullbullbullbullbull Estimating Stock Inventories bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Estimating Supplies Stock and

Yield Models Estimating Financial Returns

Plantation Models bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Fuelwood Production Data bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Fuelwood Market Prices bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Fuelwood Relative Prices bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Fuelwood Economic Values bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Plantation Costs bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

D Transport Costs and Market Structures bullbullbullbullbullbullbullbullbull E Charcoal bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

Production Processes and yields bullbullbullbullbullbullbullbullbullbullbullbullbull Charcoal Prices and Other Databullbullbullbullbullbullbullbullbullbullbullbullbullbull

F Agricultural Residues bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Residue Supplies and Energy Content bullbullbullbullbullbullbullbullbull Availability and Economic Costs bullbullbullbullbullbullbullbullbullbullbullbullbull Pellets and Briquettes bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Densification Processes and Feedstock

Characteristics bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Energy Content and Costs bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

G Animal Wastes bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Direct Combustionbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Biogas bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull

61 61 61 61 65 65 67 67 69 70 72 73 74 74 80 82 83

85 85 86 86 87 88 88 92 92 93

93

95 97 98 98

101 102 104 107 107 109 111 112 114 117

117 120 122 122 124

- vii shy

CHAPTER V ASSESSMENT METHODS AND CASE STUDIES 126 A Objectives and Structure 126 B Data Sources 126

Demand Data and Data Sources 126 Supply Data 129

C Simple Supply-Demand Projections 132 Constant-Trend Based Projections 132 Projections with Adjusted Demand 133 Projections with Increased Supplies 136 Projections Including Agricultural Land 137 Projections Including Farm Trees 137

D Disaggregated Analyses 140 Demand Disaggregation 140 Resource and Supply Disaggregation 141

E Case Studies 143

ANNEXES 1 Typical Energy Content of Fossil and Biomass Fuels 147 2 Prefixes Units and Symbols 150 3 Conversion Factors 152 4 Glossary 155 5 Summary of Classes of Constraints for Wood Stove Designs 159 6 Procedures for Testing Stove Performance 162 7 Methods for Estimating Payback Times for Stoves 164 8 Impact of Urban Woodfuel Supplies 166 9 Stages of Soil Degradation Due to Tree Loss and Removal

of Crop Residues in Ethiopia 169

BIBLIOGUPHY bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull 173

TABLES 11 Example of Energy Production-Conversion-Consumption

Stages Kerosene for Cooking bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 7 12 Primary and Delivered Energy Consumption and

Efficiencies for Three Types of Cooking Devices bullbullbullbullbullbullbullbullbullbull 20 13 Specific Firewood Consumption for Clay and Aluminum Pots bullbullbull 24 21 Estimates of Average Per Capita Biomass Fuel

Consumption in Ethiopiabullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 32 22 Annual Per Capita Consumption of Rural Household Energy

and Woodfuels Country and Regional Averages and Ranges bullbull 34 23 Per Capita Rural Consumption of Household Energy

and Biomass (GJ) Local Averages and Ranges bullbullbullbullbullbullbullbullbullbullbullbullbull 35 24 Per Capita Urban Consumption of Household Energy

and Biomass (GJ) Local Averages and Ranges bullbullbullbullbullbullbullbullbullbullbullbullbull 36 25 Fuelwood Collection Times bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 38 26 Collection Rates for Firewood bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 41 27 Cooking Fuels Used in Urban Households bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 46 28 Relationships between Energy Income and Household Size bullbullbull 49

- viii shy

29 Household Budget Shares for Energy in Urban Areas bullbullbullbullbullbullbullbullbullbull 50 210 Relative Prices of Woodfuels in Selected Countries bullbullbullbullbullbullbullbullbull 51 211 Household Energy Patterns and City Size India 1979 bullbullbullbullbullbullbull 56 212 Fuel Shares for Cooking and Heating by Income

India 1979 and 1984 bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 57 213 End-Use of Energy for Cooking and Heating in Rural Mexico bullbull 58 31 Specific Fuel Consumption for Cooking Staple Foods bullbullbullbullbullbullbullbullbull 62 32 Specific Fuel Consumption for Cooking bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 63 33 Fuel Consumption Relative Efficiencies and Cooking Times

for Different Meals and Types of Cooking Appliances bullbullbullbullbullbull 64 34 Factors Affecting Cooking Efficiencies bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 66 35 Average Cooking Efficiencies for Various

Stoves and Fuels bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 69 36 Generalized Stove Cost Index bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 71 37 Efficiencies and Total Costs of Various FuelStove

Combinations in Thailand bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 72 38 Lighting Standards for Various Household Activities bullbullbullbullbullbullbullbull 74 39 Household Kerosene Consumption for Lighting bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 75 310 Energy Use for Lighting in Electrified and

Non-Electrified Households India 1979bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 76 311 Technical Characteristics of Lighting FuelLamp

Combinations bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 77 312 Lamp Costs bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 78 313 Technical Characteristics and Costs of Electric Lighting

Technologies bull bull bull bull bull 79 314 Payback Analysis for 16 WFluorescent Lighting

Compared to 40 W Incandescent Bulbs bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 80 315 Electricity Consumption by Appliance Ownership Fiji

and Sri Lanka bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 82 41 Potential Benefits of Rural Tree Growing bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 91 42 Example of Stock and Yield Estimation Method Natural

ForestPlantation (Hypothetical Data) bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 94 43 Example of Financial Discounted Cash Flow

Method Plantation bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 96 44 Characteristics of Various Fuelwood Species bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 97 45 Retail Fuelwood Prices in Various Developing Countries bullbullbullbullbull 99 46 Relative Costs of Cooking in African Countries 1982-83 bullbullbullbull 100 47 Comparative Prices of Household Cooking Fuels in Nigeria bullbullbull 101 48 Selected Fuelwood Projects Financed by the

World Bank Since 1980 bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 103 49 Woodfuel Transport Costs General Formula and Example bullbullbullbullbull 106 410 Yields and Conversion Factors for Charcoal

Produced from Wood 108 411 Preferred Wood Feedstock Characteristics for

Charcoal Production 110 412 Retail Prices of Charcoal in Selected

Developing Countries bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 111 413 Residue-to-Crop Ratios for Selected Crops bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 112 414 Calorific Values of Selected Agricultural Residues bullbullbullbullbullbullbullbullbull 113 415 Results of Long-Term Manuring Trials in India bullbullbullbullbullbullbullbullbullbullbullbullbullbull 116

- ix shy

416 Characteristics of Various Residue Feedstocks for Densificationbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 118

417 Characteristics of Densification Processes and Products bullbullbullbull 119 418 Average Net Heating Values and Costs of

Briquetted Residues bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 120 419 Production Cost Estimates for Commercial Scale Crop

Residue Briquetting in Ethiopiabullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 121 420 Manure Production on a Fresh and Dry Basis for

Animals in Developing Countries bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 123 Cooking Energy Demand Analysis Data Needs Methods 51

and Problems 128 52 Woodfue1 Resources and Supplies Data Needs Methods

and Problems bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 131 53 Constant Trend-Based Projection Wood Balancebullbullbullbullbullbullbullbullbullbullbullbullbull 133 54 Basic Projection Adjusted for Demand bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 135 55 Basic Projection Adjusted for Demand Wood Balancebullbullbullbullbullbullbull 136 56 Projection Based on Expansion of Agricultural Land bullbullbullbullbullbullbullbullbull 138 57 Population and Fuelwood Data by Land Type Averages

for East Africa 1980bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 142 58 Household Woodfue1 Use in Urban and Rural Centers

of Madagascar bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 143 59 Contiguous Forest Cover by Province Madagascar 1983-84bullbullbull 144 510 Woodfuel Demand and Supply Balance by Region

Madagascar 1985 bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull 144 511 Projected Supply-Demand Balance for Household Energy

Antananarivo Madagascar 146

INTRODUCTION

Household energy has received increasing attention in recent years as the importance of the household sector in the energy balances of developing countries has become better understood and the problems of maintaining adequate supplies of household energy in many of these countries have become more critical Still information on household energy remains relatively scarce interpretations of the data vary widely and few non-specialists are familiar with the basic approaches to household energy analysis This handbook is intended to assist in the understanding of household energy issues by presenting a standard framework for measuring and analyzing information on supply and demand in the sector However it is not exhaustive and does not pretend to provide the last word on a rapidly changing field of knowledge Instead it is intended to serve as an interim guide and reference tool for practitioners and analysts to be revised and updated as the state of the art changes

The Importance of Household Energy in Developing Countries

Recent declines in international oil prices have reduced public interest in energy problems and have shifted the focus of national planning to more topical concerns However the economic and social costs of supplying energy in developing countries remain high and the household sector in particular continues to pose major energy problems for many countries Data from more than fifteen UNDPWorld Bank country assessment reports show the household sector accounting for 30 to 99 of total energy consumption The highest proportions are found in poorer countries where households depend almost exclusively on traditional fuels 11 the supplies of which are rapidly dwindling in many countries Thus while declining oil prices have eased the pressures of energy demand in the industrial sectors these pressures continue to grow in the household energy sector

As industrialization occurs and incomes rise the proportion of total energy used by households declines to around 25-30 as in the OECD and higher income developing countries At the same time urbanization and higher incomes lead to rapid growth in household consumption of

11 Traditional fuels refers to firewood charcoal crop residues and animal wastes These are sometimes termed biomass fuels or biofuels They may be bought and sold (commercialized monetized) or gathered without financial payment from the environment Other energy sources including coal coke kerosene liquified petroleum gas (LPG) natural gas and electricity are referred to collectively as modern or non-traditional fuels

- 2 shy

petroleum electricity and other modern fuels For example in most developing countries the growth of electricity use by households exceeds 10-12 a year and in a few growth rates have exceeded 25 a year Households are therefore a major contributor to the crises of capital skills and foreign exchange deficits which beset many developing countries as they struggle to match their energy supplies to increasing demand

Despite these trends traditional fuels still playa vital role in most developing countries and will continue to do so for the foreseeable future Some two billion people who depend on wood and other traditional fuels for their basic energy needs are facing a deepening crisis of energy scarci ty as local resources are depleted and the more distant forests are cut down The implications of this crisis reach far beyond the supply of energy itself As trees are lost and people are forced to burn fuels that are taken from the fields the land which provides their livelihood and feeds the nation may become increasingly vulnerable to erosion and soil degradation In some arid areas of the developing world this process has reached its terminal stages where the land produces nothing and starvation or migration are the only alternatives

Recognizing the severity of the fue1wood crisis the World Bank has increased the number of its projects dealing with social forestry improved cooking stoves charcoal production and other aspects of biomass utilization The direct linkage that exists between household energy consumption patterns and depletion of forest resources loss of soil cover and other environmental problems makes the analysis of household energy issues essential in evaluating these problems as well This handbook then reflects the World Banks increasing concern with these issues and its commitment to strengthening its analytical capabilities for dealing with them

Characteristics of Household Energy

Compared with industry and commerce the household sector has energy demand and supply characteristics which make assessment and project analysis at times difficult and unique There are several critical differences between the household sector and other sectors

First the household sector consists of many individual users who live in a great variety of energy landscapes There is enormous diversity in the availability and costs of energy supplies in the levels of consumption and mix of fuels employed in end-uses such as cooking water heating space heating and lighting and in technologies and energy-related preferences and modes of behavior

- 3 shy

Second most household energy use is not recorded by supply agencies but must be ascertained through household surveys This is so for the traditional fuels which dominate the household energy sector in most developing countries since they are either collected or traded outside the monetary economy or bought and sold in a mUltiplicity of small markets It is also true for anything but the most aggregate level of consumption for petroleum fuels such as kerosene and liquified petroleum gas (LPG or bottled gas) which are also bought at a myriad of retail outlets Only with electricity and piped gas are there central ized and disaggregated records of household consumption because these supplies are metered and billed

Third traditional fuels especially in rural areas represent only one aspect of the complex interrelated systems for producing exchanging and using biomass materials of all kinds including for example human food animal fodder timber and crop residues for construction materials as well as fuels Energy problems and solutions must almost invariably be considered within this total context At the same time there are no established market mechanisms in rural areas to bring supply and demand for traditional fuels into balance so that in many instances the depletion of biomass fuel resources continues unabated with severe impacts on other parts of the biomass system and on present and future household energy supplies These impacts are usually most severe for the rural and urban poor who are least able to adapt to the increasing scarcity and rising cost of resources

Fourth traditional household fuels and technologies for their use are often difficult to change largely because alternatives are not known there is no capital available to make use of alternatives and households tend to prefer to continue with age-old customs

These characteristics make it especially difficult to gather and assess basic energy data on the household sector Furthermore energy supply and demand patterns are location-specific They normally vary considerably by region district village and town and by household classes within towns National energy studies must reflect these differences if they are to provide a valid basis for planning Therefore these studies require a high degree of spatial and social disaggregation which is extremely time-consuming and costly The alternative of generalizing to the national or regional level from a few detailed surveys in some places may be quite misleading unless the survey sites are known to be representative Such detailed studies are also time consuming Consequently there is a general lack of reliable energy data for the sector and in particular of comparable data for different time periods which can illuminate trends in energy demand and supply over time

- 4 shy

Purpose of the Handbook

The major purpose of this handbook is to assist those involved in energy demand or supply planning national energy assessments or project design for the household sector To do this the authors have brought together from developing countries data on household energy consumption resources and technologies and wherever possible put them into a consistent framework This has been a challenging task partly because of the diversity of inputs mentioned above and also because of the prevalence of unreliable or incomplete data Although many bits and pieces of sound energy information exist they are scattered through a vast literature and are often expressed in such a way that comparisons and integrations are difficult or impossible unless the information is reworked altogether The Handbook is thus intended to provide a set of reference tools for conducting household energy analysis and guidance on where to find this information and how to use it in energy assessments and project design Before discussing these issues two cautions are noted

First the extreme diversity of household consumption and supply patterns usually means that truth can only be found at the local level Generalizations from these situations may often be necessary but one should always recognize that they can be at best risky and at worst downright misleading Consequently the patterns and data described in this book are no more than signposts for what to look for in particular locations

Second energy studies often fail to reach behind the facts to the underlying questions and relationships Why for example dont people plant trees when firewood is scarce and its collection takes up many hours a week Who is able to respond to fuelwood scarcity Are energy demands the main cause of tree loss Unless such questions are examined carefully in each location where action is contemplated that action will most probably fail Over the past decade the experience of energy policies and projects that attempted to address the needs of families in developing countries has not been altogether a beneficial one Project failures often can be traced to a lack of understanding of local conditions and the way people see their own priorities and options for action

Organization of the Handbook

The Handbook is divided into five sections Chapter I discusses basic energy terms and principles critical to understanding the energy units definitions data and calculations presented in the following chapters Chapter II describes household energy consumption patterns and their dependence on key variables such as income urbanshyrural location and household size Chapter III takes a close look at

- 5 shy

the end-uses of energy and the technologies which provide such services as cooking heat lighting refrigeration and space heating This initial focus on demand emphasizes the fact that energy supplies are required only to satisfy personal needs and that families frequently respond both to demand and supply options in intensely personal ways

Chapter IV examines household energy resources and supplies focusing almost entirely on traditional biomass fuels including tree growing and firewood charcoal crop residues and animal wastes Nonshytraditional energy sources such as petroleum products and electricity are not discussed since there is a vast and easily available literature on these topics

Finally Chapter V provides examples of simple assessment methods and case studies to illustrate ways in which household energy data can be put to work in energy economic and technical assessments and to warn of some methodological pitfalls

- 6 shy

CHAPTER I

ENERGY MEASUREMENT AND DEFINITIONS

A OBJECTIVES AND STRUCTURE

This chapter explains and compares the main conventions of energy measurement in general use paying particular attention to the traps and ambiguities which lie in wait in energy reports surveys and statistics Although experienced energy analysts may be familiar with much of the subject matter they are advised to skim through the chapter to ensure that they understand which conventions are used in later chapters

Section B below describes general measurement systems and discusses key definitions and terms of energy analysis It also provides basic methods for adapting the definitions for ones preferred system of measurement Section C focuses on some major analytical problems associated with end-use technologies such as cooking stoves and lighting equipment especially with measures of efficiency and utilized energy Section D provides a brief guide to basic statistical techniques for assessing the validity of survey data

B BASIC MEASUREMENT CONCEPTS

Measurement Systems and Reference Data

The System International (SI) and British system are the most coamonly used physical measurement systems This book uses the SI system as it has been adopted by most international agencies and many developing countries as well

Production and Conversion Systems

All use of fuels (including electricity) involves a series of energy conversions as shown in Table 11 Usually these conversions change the physical nature of the fuel or the form of energy in order to increase its utility An example is the conversion of crude oil into kerosene followed by the conversion of kerosene to heat in a cooking stove and finally into cooked food Invariably some energy is lost to the environment during these conversion processes

This concept is basic to energy measurement and to such important factors as the energy content of fuels and the efficiency of conversion processes However by comparing different stages in the production-conversion chain one can derive various definitions and

- 1 shy

Table 11 Example of Energy Production-Conversion-Consumption Stages Kerosene for Cooking

General Form of Term for Fuel or Conversion Stage Energy Technology Comments

A Resources Reserves

Recoverable Reserves

B Primary Energy ~

C Secondary Energy

D Delivered Energy ~ (heat of combustion)

E Util ized Energy ~ for Cooking (PHU or heat uti I ized

Crude oi I in ground

Crude oi I extracted

Kerosene

Kerosene (purchased by household)

Heat absorbed by cooking food etc (cooked food)

Production well

Refinery

(Distribution amp

Marketing)

Cooker and cooking pot etc

Estimates uncertain

Varies with finds technology costs

Energy use losses (eg gas flaring)

Energy use losses

Delivered energy minus heat escaping around cooking pot radiation losses from stove body etc See Figure 15

These terms are the most commonly used

measures of these important values Care therefore must be taken to use consistent definitions and to appreciate what definitions others are using before applying their results To illustrate these points Table 11 presents a simplified chain for the production of crude oil its conversion to kerosene and the use of kerosene in cooking The terms used in this book for each stage are given in the first column Some comments on each may be useful

Resources and Reserves have various subdivisions to indicate the certainty of the estimates or the availability of reserves under different technological and economic conditions For fuels such as oil gas and coal the meaning of these terms is usually indicated clearly in reserve assessments

- 8 shy

Primary Energy is sometimes called Primary Production (UN) Total Energy Requirement (OECD) and Gross Consumption (EEC) It measures the potential energy content of the fuel at the time of initial harvest production or discovery prior to any type of conversion It is often used for recording the total energy consumption of a country which is misleading because it ignores the conversion efficiencies at which the fuel is used

Secondary Energy is sometimes called Final Energy (EEC OECD) It differs from Primary Energy by the amount of energy used and lost in supply-side conversion systems such as oil refineries power stations biomass gasifiers and charcoal kilns

Delivered Energy is sometimes called Received Energy since it records the energy delivered to or received by the final consumer such as a household Examples are domestic kerosene purchases and firewood as collected and brought to the doorstep II In most energy statistics Delivered and Secondary Energy are the same for fossil fuels and electricity because Secondary Energy is estimated from sales to final consumers (ie Delivered Energy) Any (small) losses incurred in distribution and marketing are therefore included in the conversion from Primary to Secondary Energy

Util ized Energy is sometimes called energy output end-use delivered energy or available energy The term utilized is the most appropriate because we are measuring the amount of work or utilized heat to perform a specific task or service The provision of these services is the ultimate purpose of the entire energy production and conversion system Utilized energy may be as little as 5-8 of delivered energy with an inefficient conversion technology such as an open cooking fire or as high as 95-100 of delivered energy in the case of electric resistance space heating

Since utilized energy records the utility to the consumer of his or her consumption of fuel for any desired task it is frequently used as the basis for comparing fuel prices (eg dollar ($) per MJ of utilized heat for cooking) and for examining the economics and energy savings due to fuel and technology substitutions (eg switching from open cooking fires to closed stoves)

However the concept of utilized energy is sometimes difficult to apply For example if a cooking fire provides multiple end-use services--such as space heating and lighting as well as heat for cooking--it is neither practical nor sensible to try to measure the utilized energy for each service The same is true of lighting where the distance from the light source to the user and the quality of light output (ie the spectral range) is at least as important to the amount of energy used or the consumers motivations to switch technologies as any measure of utilized energy For these reasons it is often better to consider energy use and compare technologies in terms of specific fuel consumption for a particular task or time period eg the amount of

- 9 shy

cooking fuel per standard meal or weight of staple foods or the kWh of lighting electricity per household per day These issues are discussed further in Section C

Measurement Units

Four basic types of units are used in energy measurements and assessments

Stock energy units measure a quantity of energy in a resource or stock such as the amount of oil in a reserve kerosene in a can or wood energy in a tree at a given point in time Examples are tons of oil equivalent or multiples of the Joule (MJ GJ PJ) Although stocks may appreciate or decline over time these changes are often most usefully given as stock units eg for a growing fuelwood plantation as the standing stock in units of weight or energy equivalent at the start of one year and of the following year

Flow or rate energy units measure quantities of energy produced or consumed per unit of time and are used for Primary Delivered and Utilized Energy consumption Examples are million barrels of oil per day (MBD) PJyear or MJday of cooking fuels Frequently the time unit is omitted as when a countrys (annual) primary energy consumption is given as so many million tons of oil equivalent TOE These units are the same as power units eg kilowatts (kW)

Specific energy consumption relates a quantity of energy to a non-energy value It is often referred to as an energy intensity Examples are MJ per kg of cooked food or MJ per unit of household income (MJ$)

Energy content or heating value measures the quantity of energy in a fuel per unit weight or volume Examples are MJkg and MJlitre

Gross and Net Heating Values

The heating value (HV) of fuels is recorded using two different types of energy content--gross and net Although for petroleum the difference between the two is rarely more than about 10 for biomass fuels with widely varying moisture contents the difference can be great Unfortunately the basis on which HVs are recorded is often omitted and one frequently finds both methods used for different fuels in the same report or energy survey

Gross Heatin~ Value (GHV) sometimes erroneously referred to as higher heating value refers to the total energy that would be released through combustion divided by the weight of the fuel It is used in the

- 10 shy

energy statistics of the United Kingdom the USA and many developing countries and in many household energy surveys

Net Heating Value (NHV) sometimes called the lower heating value refers to the energy that is actually available from combustion after allowing for energy losses from free or combined water evaporation It is used in all the major international energy statistics (UN EEC OECD) Net values are strongly recommended and are used throughout this book

The NHV is always less than GHV mainly because it does not include two forms of heat energy released during combustion (1) the energy to vaporize water contained in the fuel and (2) the energy to form water from hydrogen contained in hydrocarbon molecules and to vaporize it A simplified view of the combustion process should clarify this difference

Combustion Process Outputs

1 bull Heat NHV

2 Hot water vapor formed from hydrogen including its latent heat of vaporization GHV

Fuel + Air Combustion

3 Hot water vapor from contained water Including latent heat

4 Carbon Dioxide and monoxide Nitrogen OXides etc

1 = NHV Note 1+2+3+4 bull GHV

Clearly the difference between NHV and GHV depends largely on the water (and hydrogen) content of the fuel Petroleum fuels and natural gas contain little water (3-6 or less) but biomass fuels may contain as much as 50-60 water at the point of combustion It is also fairly obvious that few household combustion appliances can utilize the outputs labeled 2 3 and 4 Consequently on a net basis the energy value of a fuel reflects the maximum amount of heat that normally can be obtained in practice (ie output 1) On a gross basis the energy value overstates this quantity by the ratio GHVNHV or (Outputs 1+2+3+4)

Output 1

- 11 shy

Heating Values and Moisture Content

Annex 1 presents typical NHVs for the most common solid liquid and gaseous non-biomass fuels With solids there can be large variations in heating value due to differences in water ash and volatile content Liquid fuels have a much more uniform energy content but there are still slight differences due to refinery specifications and blending etc Local values should be used if possible otherwise the data in Annex 1 can be used for reasonable approximations In any analysis particularly when dealing with wet fuels the energy contents (NHVs) employed should be recorded clearly

For biomass fuel s special care must be taken to measure and record the water (moisture) content wherever possible The moisture content can change by a factor of 4-5 between initial harvesting and final use and is critical both to the heating value on a weight or volume basis and to differences between GHV and NHV This section aims to clarify these concepts and provides conversion factors for the commonly used measures

Moisture content can be given on a wet or dry basis The basis should always be specified (although many reports omit this necessary information) Moisture content dry basis (mcdb) refers to the ratio of the weight of water in the fuel to the weight of dry material Moisture content wet basis (mcwb) is the ratio of the weight of water in the fuel to the total weight of fuel 80th are expressed as a percentage The respective formulae are

Moisture content () Water weight in fuel x 100 Dry basis (mcdb) = Dry weight of fuel

Moisture content () Water weight in fuel x 100 Wet basis (mcwb) Water weight + dry weight of fuel

Water weight in fuel x 100 = Total weight of fuel

To convert between wet (W) and dry (D) basis the following formulae are used

W= D(l + D100) D = W(l - W100)

This relationship between the several heating value definitions is graphically represented in Figure 11

Heating values of biomass fuels are often given as the energy content per unit weight or volume at various stages green airshydried and oven-dried material They correspond to the following

- 12 shy

FIGURE 11 Relationship between Several Heating Value Definitions

Mass (kg) Energy (MJ)---r------i-shyCombustible

Fiber

Ash

Water

-

~

Net D

High E

DryG

Wet BWater

A losses

F Water

World Bank-307366

HEAT VALUE FORMULAE

High (Over-dry) Heating Value = o (MJ) E (kg)

o (MJ)Gross Heating Value = 0 E + F A (kg)

C (MJ)Net Heating Value = C E + F A (kg)

MOISTURE CONTENT FORMULATE

F F x 100 Moisture Content wet Basis = E + F G

Moisture Content Dry Basis = F x 100 E

- 13 shy

Green refers to the living plant or the plant at the point of harvest

As received refers to the moisture content at a given point in the fuel chain

Air-dried refers to the stage after the fuel has been exposed for some time to local atmospheric conditions ie at any stage from harvesting to the conversion of the fuel either to another fuel or by combustion to heat energy

Oven-dried means that a fuel has zero moisture content and is sometimes referred to as bone dry

Moisture contents of green and air-dried wood will differ depending on several factors including (1) the species (2) atmospheric humidity and hence climatic and seasonal factors (3) drying time and (4) drying conditions including temperature and ventilation In the humid tropics green wood may typically have a moisture content of 40shy70 mcwb After prolonged air drying this value will fall to 10-25 mcwb depending on atomospheric humidity (See Figure 12) Since many families keep a short-term stock of wood in the kitchen and often close to the cooking fire further drying may occur to give moisture contents as low as 10-20 mcwb Typical values for the moisture content of wood as burned are in the 7-15 mcwb range However substantially higher moisture contents are found in zones or seasons of heavy rainfall andor where wood is scarce so that the air-drying time between cutting and burning is reduced to only a few days (and in exceptional cases as little as 24 hours)

FIGURE 12 Effect of Relative Humidity on Equilibrium Mositure Content of Wood

25

30

~ ~ 20 11

2 a

15 ic 0

15 ~ ~ ~ 8 u u i

10 J

~ ~ middot0

o 20 40 60

5

Relo1lve Humidity ()

Source Sham (1972) World 8ank-307367

- 14 shy

The oven-dry (aD) heating value is an unambiguous measure of the energy content of the combustible material in solid fuels and 18

frequently given in reference data [FAa 1983c OTA 1980J It is determined in the laboratory by weighing a sample before and after it 1S

dried in an oven until the weight no longer changes so that one can assume that all moisture has been driven off and then measuring the heating value of the dried sample

The procedure for converting the oven dry gross heating value to net heating value or gross heating value for any moisture content is fairly simple and accurate Considering a 1 kg piece of wood containing W kg of water the weight of oven-dry combustible material plus ash etc is (l-W) kg Suppose that the oven dry gross heating value of this material is Z MJkg Then the gross heating value of the wood sample is Zl-W) MJkg For the net heating value we must deduct the heat energy for the hydrogen water and free water Most oven-dry woody materials contain close to 6 of hydrogen by weight which would correspond to a hydrogen term of 13 MJ per kg dry material or 13 (l-W) for the sample For the free water a value of 24 MJkg is frequently used The water term is thus 24 (W) The net heating value of the wood sample in SI units (MJkggt is therefore zl-W) - 13 (l-W) - 24 (W) This reduces to Z - 13 - WZ+ll)

To summarize in 81 units of MJkg the conversion formulae are

NHV wet basis = Z-13 - (WlOO) (Z + 11) NHV dry basis = (lOOZ - 130 - 24D) (100 + D) GHV wet basis = zl - WlOO) GHV dry basis = Z (l-DlOO + Draquo

where Z is the oven-dry gross heating value and Wand Dare the percentage moisture contents on a wet and dry basis respectively

For easy reference these values are plotted against moisture content in Figure 13 using a reference wood of 20 MJkg oven-dry gross heating value

This reference value is a reasonable first order approximation in the absence of actual measurements Tests on 111 species of tropical fuelwoods from Africa Asia and South America obtained an average of 200 MJkg (oven-dry Gav) with a standard deviation of under 06 MJkg or less than 3 of the mean value [Doat and Petroff 1975] The lowest value was 184 MJkg and the highest 220 MJkg These differences are less important than variations due to moisture content as Figure 13 makes clear However some fuelwoods with a high ash or silica content such as bamboo and coconut have lower values of about 17 MJkg (oven-dry GHV) while resinous woods such as the American pine species have values in the 24-28 MJkg range

- 15 shy

FIGURE 13 Heating Values for Wood as a Function of Moisture Content (for reference wood of 20 MJkg oven-dry gross heating value)

Heating Value

(MJkg)

20

GHV

NHV

I I MCWBo

I 30 40 60 80 100 MCDB

World Bank-307368

10 20

These values refer to large pieces cut from the trunk or main branches For small branches and twigs which are widely used as fuels by the poor heating values tend to be both lower and more variable than for stemwood from the same species Typical values are not as well recorded as they are for stemwood but one series of tests in South India found a mean value of 174 MJkg (oven-dry GHV) for 15 species with a standard deviation of only 02 MJkg [Reddy 1980]

However it is a reasonable practice to use 20 KJkg oven dry if no original measured data are available for the wood concerned and there is no basis for believing that a markedly lower or higher value obtains If the design of combustion systems is involved then actual heating values should be obtained through laboratory analysis

- 16 shy

Volume Density and Moisture Content

Fue1wood resources production and consumption are often reported in volume terms This is the usual practice among foresters since timber is normally sold in units of volume -- usually as the actual (or solidtt

) volume of the wood Frequently and especially in informal markets and household surveys the only record of fue1wood quanitites produced sold or consumed is a volume measure based on the outer dimensions of a loose stack or load containing air spaces between the wood pieces such as the stere cord truckload headload or bundle

To use such measures for energy analysis two approaches can be taken The first is to convert stacked volume to a weight and then proceed as outlined above This can be done for small loads by weighing a number of samples with a spring balance or for a large load (eg truckloads) by use of a weighbridge The second approach is to convert stacked volume to solid volume This can be done for small loads by immersing them in water and measuring the volume of water displaced If direct measurements are impractical local conversion factors or rules of thumb must be used these are usually known by foresters fue1wood truckers wholesales and retailers etc No general guidelines can be given here since both conversions (stacked volume to weight stacked volume to solid volume) vary greatly by location

If it is not possible to convert volumes to weights for energy analysis the volumes of fuels have to be converted to a volumetric measure of energy content To do this a series of three conversions is often required These are described below However one should first note that the basic density and the specific gravity of wood are always reported on an oven-dry basis For consistency the conversion formulae are based on weights in kilograms (kg)

1 Conversion of oven-dry volume to oven-dry weight

Oven-dry weight (ODW) (kg)

= Vo~ume (m )

x Basic density (kgm3)

and since

Basic densisecty = Specific gravity x 1000 (kgm ) of dry matter (gmkg)

3(gmcm 1 (kglton) (tonsm )

then

Oven-dry weight (ODW) = Volume x Specific gravity x 1000

- 11 shy

2 Conversion of oven-dry weight to actual weight for specific moisture content

Actual wet weight = oow(l-wlOO)

where W is the percentage moisture content wet basis (mcwb)

3 Conversion of actual wet weight at specific moisture content to net heating value given the oven-dry value

Use actual weight and the formulae given on page 14 for heating value per unit weight These formulae can be combined to give a single formula for converting

from Volume (V) basic density (80) oven-dry gross heating value (Z) and percentage moisture content wet basis (W)

to the net heating value (NHV) as recommended and used in this book

NHV = V x 80 x (Z - 13 - (WlOO) x (Z + 11raquo (of given volume) 1 WlOO

3where volume is in m weight is in kg and energy is in MJ

The critical importance of correctly applying all the concepts discussed above deserves illustration with an actual example of a fuelwood production and delivery chain

3The starting point of the chain in this example is one solid mof green wood at the point of harvest weighing 12658 kg (See Figure 14) The basic density of the material is 06 (600 kgm3) and the ovenshydry energy value is 20 MJkg The moisture content (~cwb) is 526 Consequently the volume of combustible material is one m and its weight 600 kg

The wood is air-dried in two stages between harvesting (primary energy) and its purchase by a household (delivered energy) and between this stage and its use in a cooking fire (delivered energy at the point of use) Figure 14 records at each stage the values of volume weight moisture content actual density and total energy measured in gross and net heating values (GHV and NHV) - shy

As one would expect since water is lost between each stage the weight density and moisture content decrease progressively 2 However this is not so for the net heating value or for the total energy content of the sample on an NHV basis

Volume also decreases slightly with drying by about 5 in the example shown (FAO 1983 c] Figure 14 assumes a constant volume

- 18 shy

FIGURE 14 Changes in Physical Quantities during States of Air-Drying Fuelwood

Water loss 4658 kg

Water loss 13333 kg

Water ~ 6658 kg Water

________________________~-----~~~~------~w~a~~-r----~ ~-----66-6-7-k-9----~ 200 kg

I-

CombustionCombustion MaterialMaterial 600 kg 600 kg

Combustion Material 600 kg

World Bank-307369

Point of Use Del ivered

(point of use)

approx 1 66667 approx 66667

10 111

12000 11060 (1659)

ENERGY STAGE

Volume (m3) Weight (kg) Density (kgm3) Moisture content (mcwb)

Content (lIcdb)

TOTAL ENERGY (MJ) GHV basis NHV basis

(NHV MJkg)

Basic Data

Harvest Primary

12658 12658

526 111

12000 9620

(750)

Basic density

Point of Sale Delivered

approx 1 800

approx 800

25 333

12000 10744 (1343)

600 kgm3

Oven-dry gross heating value 20 MJkg

On a GHV basis both the heating value (MJkg) and the total energy content of the sample (MJ) remain constant

Using a NHV basis the heating value and the total energy content of the sample increase This is~not a case of creating energy out of nothing since the energy content in question refers to the heat that can be usefully extracted from the fuel in a device such as a

- 19 shy

cooking fire This is so much greater per unit weight for dry wood than wet wood that it more than compensates for the loss of weight due to drying

C UTILIZED ENERGY EFFICIENCY AND SPECIFIC FUEL CONSUMPTION

The delivered energy content of a fuel measures the potential heat available from it When the fuel is used for a specific end-use task such as cooking food only a fraction of this energy is usefully employed for that task This quantity is called the utilized energy (for that specific task) The fraction of the energy utilized defines the efficiency of the end-use device (for that task) Efficiencies are usually defined in terms of delivered energy but can also be given on a primary energy basis In the first case

Efficiency for task (Delivered Energy basis)

= Energy utilized for task Energy delivered to conversion device for task

For household applications stove or appliance efficiency is commonly referred to This is the utilized energy efficiency expressed as Percentage Heat Utilized (PHU)

This seems simple enough However few energy conversion devices--least of all cooking fires and stoves plus cooking equipment-shyare simple in terms of their energy flows Still less are they simple in the way in which people use them The critical importance of correctly measuring efficiency and utilized energy for the household sector demands that we examine these concepts carefully

Primary and Delivered Energy Efficiencies

This topic is relatively simple It is demonstrated in Table 12 which compares the primary and delivered energy requirements of a wood fire a kerosene stove and an electric cooker which perform the same task of providing 10 units of utilized energy for cooking

The table shows that although the electric cooker has the highest delivered to utilized efficiency it has the lowest primary to utilized efficiency and hence consumes the most primary energy of the three cooking methods If electricity is generated from oil more oil would be consumed than with the kerosene cooker For the consumer it is the delivered to utilized energy efficiency that matters since this determines the energy cost for the task ie delivered energy (KJ) x unit price ($KJ)

-~~----------------------

- 20 shy

Table 12 Primary and Delivered Energy Consumption and Efficiencies for Three Types of Cooking Devices

Wood Kerosene Electric Fire al

= Stove Cooker

Primary energy (PE) ~ 67 37 56

Conversion efficiencl Primarl to Delivered ~ 115

(air drying) 09 (refinery)

030 (generation)

De livered energy (DE) ~ 17 333 167

Conversion efficiencl Del Jvered to Uti I ized =UEIDE Utilized energy (UE) ~

013

10

030

10

060

10

Conversion efficiencl Primarl to Util ized UEPE

015 026 014

a Energy values in units to cook an arbitrary unit quantity of food b Excludes transmission and transport

Definitions of Efficiency

When fuel is burned its energy is usually transferred to the end-use task in several stages Energy losses of various kinds occur on the way Measures of efficiency and utilized energy therefore depend critically on the stage at which the heat flow is measured for example with a cooking stove and pot whether one measures the heat from the stove opening the heat absorbed by the pot or the heat absorbed by the food

This point is illustrated in a highly simplified way in Figure 15 In practice the energy flows and losses are much more complex than this so that it is often difficult to determine what definitions of utilized energy and efficiency are being used when different technologies are assessed Since different definitions can greatly affect the reported results efficiency and utilized energy should be used with caution Alternatively one should rely on less ambiguous measures such as the specific fuel consumption of a particular end-use appliance and task ie a measure of the fuel actually used for a process such as cooking a particular foodstuff or meal in the actual environment where some intervention is planned

- 21 shy

FIGURE 15 Energy Losses during Cooking With a Stove and Pot

+--------Losses In Hot Water Vapour from Pot

Contents (E)

~---TI-6-r-iIiq--------Heat Transfer Loss Pot

+~q~te~I-------- to Food (D)I- Heat Transfer Loss Stove 10 Pol (C)

ftt--------- Heat Transfer Loss through Equipment (B)

utJ)~If-t--------- Combustion Efficiency Losses (Al

World Bonk-30736 10

In order to compare technologies (see Chapter III) some distinction has to be made between the various measures of efficiency In this book three basic terms for efficiency are used ~

a Combustion Efficiency allows for energy losses in the combustion process and heat that does not reach the point where it could in theory be transferred to the the final task (eg A and B in Figure 15)

Combustion Efficiency Heat Generated by Combustion (MJ) Del ivered Energy of Fuel (MJ)

b Heat Transfer Efficiency allows for energy losses between the combustion outlet and the end-use task especially heat transfer and radiation losses (C 0 and E in Figure 15)

Heat Transfer Efficiency = Energy Absorbed by End-use Task (MJ) Heat Generated by Combustion (MJ)

c System or End-use Efficiency is the product of the Combustion and Heat Transfer Efficiencies or the overall efficiency It is often referred to as conversion gross thermal and end-use efficiency

3 One sometimes finds the terms net or Second Law efficiency in the energy literature especially in reports on household energy conservation This is a source of much confusion It refers to the thermodynamically minimum amount of delivered energy required to perform an end-use task This is invariably much less than that for any practical device Its use is not reconunended since it is of little practical value in any consideration of actual technologies

- 22 shy

d Percentage of Heat Utilized (PHU) is the energy utilized and expressed as a percentage of that available at any stage in the energy conversion process The overall PHU is commonly referred to as appliance (eg stove) efficiency

Specific Fuel Consumption Energy Intensity and Fuel Economy

The previous section discussed the difficulties in defining critical terms such as efficiency and utilized energy even in controlled laboratory tests These difficulties are greatly increased when one considers real life conditions

In real life cooks may light the cooking fire or stove well before they begin cooking They mayor may not quench the fire when cooking is finished They cook a variety of meals each using their own methods Pot lids may be left on or taken off when simmering food Equally important the cooking fire may well serve multiple purposes including space heating water heating for washing or cleaning dishes and clothes lighting or a social focus A recent survey of Maasai households in Tanzania for example found that the cooking fire was typically kept alight for about 16 hours a day with widely varying rates of combustion and fuel use in order to provide all the end-use services just mentioned [Leach 1984]

In these real circumstances estimates drawn from laboratory tests of utilized energy and end-use efficiency are of limited value Broader and looser measures based on actual observations of energy conshysumption for a class of end-use tasks should be used instead These measures include specific fuel consumption (SFC) and energy intensity Some examples are

Cooking MJ per meal MJ per person per meal MJ per kg food cooked MJ per household per day (for cooking)

Lighting MJ per lamp per day (allowing both for rate of consumption--watts liters kerosenehour--and for time period used--MJ per household per day (for lighting)

General MJ of woodfuel per household per day (used for inseparable end uses including cooking and heating)

These measures can be used for assessing changes in technology and fuel just as effectively as measures of end-use efficiency or utilized energy Of course if a more efficient technology is introduced the specific fuel consumption is likely to fall But it may not fall as expected from a direct comparison of the before and after efficiencies the users may employ the new technology in a different manner from the old one for example Only a before and after comparison of specific

- 23 shy

fuel consumption can capture such effects An example of its use in technology and fuel substitution is given below

Example Substitution of cooking pot and cooking heat source

A family cooks on an open fire using clay pots (Technology 1) The kitchen is outside the house and cooking is the only service provided by the fire Consumption of firewood is measured over a period Further measurements are made of firewood energy consumption over different periods of time when the family uses (2) an aluminum cooking pot with the open fire (3) a metal stove with a clay pot and (4) a metal stove and aluminum pot

After normalizing the consumption for Technologies 2 3 and 4 to the same time period as for Technology 1 the energy consumption levels in MJ are found to be

Consumption Technology MJ kg ~ Ratios

1 Open fire clay pot 1667 834 40 2 Open fire aluminum pot 833 417 20 3 Stove clay pot 555 278 t 33 4 Stove alUMinum pot 417 209 10

a Based on a conversion ratio of 20 MJlkg

The consumption ratios give an unambiguous reading of the re1ative fuel consumption and savings in moving from one technology to another (for this family) For example a 66 savings is achieved by switching from Technology 1 to Technology 3 Note particularly that it is not necessary to estimate either the utilized energy for cooking or the efficiencies of each technology package Indeed the relative fuel consumption for each technology option may well not be the same as the relative end-use efficiencies recorded independently of the household environment since in moving from one technology to another the family may alter its cooking methods time for cooking etc

In summary efficiency and utilized energy are basic and invaluable tools for people who are designing and developing technologies Efficiency measures are also important for comparing and marketing technologies they provide an unbiased and standarized performance yardstick for each technology--an ttenergy label They are also valuable for the energy planner and analyst when more direct data on the actual fuel consumption of real households is not available as a first order approximation one can assume that the fuel consumption of Technology A will differ from that of Technology B according to their relative end-use efficiencies (when used for the same tasks by similar

- 24 shy

classes of household However this assumption can be misleading as we shall see in Chapter III where the substitution of kerosene by electricity for lighting is discussed Wherever possible actual consumption data and the concepts of specific fuel consumption or energy intensity should be used for broad household energy assessments

D BASIC STATISTICS

Data Validity

Most quantities related to household energy use show substantial variation for example between households or in the same household from day to day Although the average (mean) of any such collection of data is a useful figure it is rarely sufficient One usually also needs an indication of the degree of certainty associated with the average This is particularly important when comparing two sets of data such as the energy consumption of a cooking stove and the traditional fire that it is intended to replace

To illustrate a typical situation where such an exercise would be desirable Table 13 below gives two sets of data on firewood use for cooking derived from field tests in 13 households in South India One set is for clay cooking pots the other for aluminum pots On average cooking with aluminum pots seems to require about two-thirds as much fuel as with clay pots the averages for each sample are 099 and 150 kg respectively However there is a large spread in consumption in each case In order to establish whether this observed difference 1S

statistically significant we would need to establish the certainty associated with the average values This is called analysis of variance and is used to test hypotheses For example the hypothesis might be that the average consumption for each type of pot is indeed different The test is then used to accept or reject the hypothesis

Table 13 Specific Firewood Consumption for Clay and Aluminum Pots

(kg wood per kg food cooked)

Predominant pot type CI Aluminum

Original data (13 measurements)

Mean weight = No of observations (N) Standard deviations (SO)

187 145 090 160 167

150 5

0367

069 197 091 068 053 141 088 085 099

8 0475

Source Geller and Dutt [19831

- 25 shy

With analysis of variance one could conclude from the above sample with 95 certainty that the average firewood consumption for a large population using clay pots lies between 105 and 195 and similarly that the 95 confidence interval for the aluminum stove would be between 060 and 138 Since these intervals overlap we cannot be 95 certain that average firewood consumption with the two types of pots is indeed different

Even if the above intervals had not overlapped we would only be able to place as much significance on the results as the reliability of the sample figures themselves In other words one should not let the mathematics produce a false sense of reliability in the conclusions beyond the reliability of the data itself

Elasticities

The use of elasticities is conunon in the household energy literature An elasticity indicates the quantity by which one (dependent) variable changes when a second (independent) variable is changed by a unit amount For example an electricity-income elasticity of 08 for the household sector indicates that domestic electricity consumption increases by 08 for each 1 increase in household income when other factors are held constant An electricity-price elasticity of -03 means that consumption falls by 03 for every 1 increase in electricity prices (other factors remaining constant) The following equation links electricity consumption to income and price using these elasticities

E = A x Ib x pc (or in the above case E = A x I Obull8 x p-Obull3)

where E 1S electricity consumption I 1S income and P 1S

electricity price A is a constant and band c are the income and price elasticities of electricity consumption respectively

The above relationship between consumption and price is known as the own-price elasticity of demand since it reflects the extent to which demand for a particular fuel would change in response to a change in its own price However because households can substitute a number of different fuels to meet their household energy needs changes in the price of a particular fuel will affect the consumption of other fuels well This effect is known as the cross-price elasticity of demand represents the percentage change in consumption of fuel A as a result a 1 change in the price of fuel B

as it of

equation We can represent this relationship mathematically by an

FA b d1 d2 d3 d7

= AI PA PB PC bullbullbull PG

- 26 shy

where A is a constant I the income level Pi the price of fuel i ~nd FA the consumption of fuel A Then b would as before represent the income elasticity of demand for fuel A and dl the own-price elasticity of demand for fuel A while d2 d3 bullbullbullbull d 7 would be the respect i ve cross-price elasticities of consumption of fuel A with respect to the prices of fuels B C bullbullbullG While dl (the own-price elasticity) will in general be negative d2 through d7 (the cross-price elasticities) will generally be positive since an increase in the price of fuel B is likely to lead to an increase in the consumption of fuel A

Studies have shown that cross-price elasticities (and therefore relative prices) are important in explaining shifting consumption patterns of the various household fuels For example a study in Syria found that contrary to what might be expected household kerosene consumption has been decreasing in recent years in the face of falling real kerosene prices (see Figure 16) [UNDPThe World Bank 1986] However during the period under question real LPG prices had been decreasing more rapidly than that of kerosene creating an effective increase in the price of kerosene relative to LPG Not surprisingly then t the consumpt ion of LPG increased over that period Thus it is important to consider the own-price and cross-price effects when analyzing the consumption patterns and projections of the various household fuels and prices

Elasticities when mathematically part of a homogeneous relationship as above can be estimated by regression of the basic data Regression methods are explained in most introductory texts on statistics

Two important measures are normally given with elasticity estimates of this kind to indicate the statistical uncertainty associated with the r~ported value The adjusted coefficient of determination (adjusted R ) measures the proportion of the variance or spread in the dependent variable explained by the independent variables and adjusted for the degrees of freedom The maximum value is 1 Thus if the r~gression of electricity consumption on income and price has an adjusted R of 09 it indicates that income and price account for about 90 of the observed differences in electricity consumption

The t-statistic indicates the reliability or statistical significance that can be placed on the reported elasticity It equals the value of the estimated coefficient ltelasticity) divided by its standard error The larger the t-statistic the more reliable is the estimate of the coefficient Roughly speaking if the t-statistic is less than 20 the coefficient has little explanatory power and should be ignored

- 27 -

FIGURE 16

Household Kerosene and LPG Consumption (Thousand Tons)

500 -----------------------------------------------

400

300

200

100

fIIII-- fIIIIfIIII

fIIII-_fIIII filii filii Kerosene

~ -shy

--------shy-

LPG ~ ~ ~ ~

~ ~

~

o ~__________________________________________~

1974

Comparison of Real Price of Kerosene and LPG (1980 SL per liter)

1984

08 r-----------------------------

07 06

Kerosene Price - I

05 - - I - I shy

- I LPG Price shy --~-- ---shy-shy --

04

03

02 ~______________________~

1974 1984

Source UNDPlWorld Bonk (1986)

World Bonk-31074

- 28 shy

CHAPTER II

HOUSEHOLD ENERGY CONSUMPTION

Households use energy for many purposes How much they consume and the types of fuel they use depend on a variety of factors These include issues of supply such as the availability of fuels and the personal or cash costs entailed in obtaining and using them But they also include many factors which can only be understood well by looking at the needs and behavior of energy consumers A major objective of this chapter is to show why an understanding of household energy must be rooted in a sensitive approach to issues of demand as well as those of supply

The second main objective is to describe and attempt to explain the enormous variety of household energy consumption patterns that is found across the developing world These patterns usually differ greatly not only between countries and national regions but even between locations only a few miles apart In most cases remedies for fuel supply and demand problems have to be based on a good understanding of local conditions and the key variables that affect the levels of demand and types of fuels that are used

Section B takes up these lssues by describing the major sources of data on household energy consumption and what they can--and cannot-shytell one about present demand patterns and their likely evolution over time

Section C examines the major variables that determine the level of household energy consumption and types of fuel used such as income rural and urban location and household size One aim of this section is to highlight the intricate and personal nature of many household energy choices

Section D gives an overview of the typical responses of rural families to increasing fuel scarcity and compares them to the reactions of urban households This provides a useful framework for considering household energy demand and supply issues

Section E provides a brief introduction to energy end-uses such as cooking heating and lighting by discussing their relative importance in total household consumption The more detailed examination of end-uses and end-use technologies is deferred to Chapter III

- 29 shy

B DATA RESOURCES

Within any country there may be four main types of data sources that provide information on household energy use and related variables Their quality varies widely and each has its own advantages and limitations

Mational Energy Balances

Most countries have energy balances which record domestic production trade conversions and losses and delivered energy consumption for the major types of non-traditional energy Usually these energy balances are developed on a regular annual basis but they may exist for only a few sample years Final consumption is broken down in greater or lesser detail by major sector Data on energy prices sometimes are included

At the present time most energy balances are based only on supply data This has two serious drawbacks for making assessments of the household sector First it is difficult from the supply side to separate household consumption from that of the commercial sector (shops hotels and restaurants artisanal workshops etc) and public sector So households are often grouped with these sectors Even if they are not they are almost invariably treated as a homogeneous unit with no breakdowns by crucial energy-related variables such as urban-rural location income or sub-region Second the consumption of traditional fuels--if they are included at all--will be very approximate As mentioned in the introduction traditional fuels are either collected from the local surroundings or traded in unofficial markets The only way to determine the quantities involved is by taking (local) surveys of household and fuel trading practices Although many such surveys have been conducted across the developing world few of them have been large enough or carefully enough prepared to provide reliable estimates of national or sub-regional consumption of traditional fuels Without such surveys national energy balances are of little value for assessing time trends in household energy use

Mational Budget Surveys

The few nationally representative surveys that have been conducted are usually undertaken by the national statistical office or finance ministry to determine the patterns of household expenditure or demographic educational and other socio-economic factors Since these are important measures for economic analysis and planning the survey samples are usually large--often around 10000-20000 households--and truly representative of regional urban-rural and income differences

National surveys are normally the only statistically valid sources of data on household energy consumption and related variables

- 30 shy

However the richness and reliability of the energy data they provide varies considerably For example

a Information is normally based on respondents recollections of expenditures over a recent period such as the preceding week With electricity and piped gas billing data is normally used so that estimates are reasonably good With all other energy sources there are obvious risks that respondents either underestimate or overestimate their expenditures If they do both equally the average for each group should be fairly reliable However there is evidence that for various reasons respondents may consistently bias their answers one way or the other 1

b Budget surveys rarely include information such as indications of fuel availability or abundance scarci ty energy prices or ownership and type of energy-using equipment Their value as tools for technical energy assessments therefore is limited

c Large nationally representative surveys are rarely conducted more frequently than every five years or so due to their high cost With each survey the range of data collected and sampling procedures may change Therefore it is rare to find consistent time series data on consumption in relation to key variables

d Budget surveys usually include expenditures on non-marketed gathered fuels by converting estimates of consumption in physical terms into cash equivalents using an imputed price These expenditures are of course imaginary Furthermore the imputed price may not be published so one cannot work back to physical quantities However this imputed price can usually be obtained from the originators of the survey

e Care must be taken 1n converting expenditure data for electricity and gas to consumption in physical units because tariff structures usually create different unit prices for small and large consumers If the tariff structure is known the conversion can be made fairly simply

1 In a survey of 180 households in Central Java people estimated how much wood they consumed Consumption was also weighed The ratio of estimatedweighed consumption ranged from 028 to 22 using average results for 32 sub-groups based on village and household size Yet the ratio for the whole sample was 095 or very close to unity (Kuyper and Mellink 19831 This balancing out of individual differences is not found in all surveys and should not be relied on

- 31 shy

National Energy Surveys

In some countries (or provinces states etc) relatively large representative surveys have been conducted specifically to measure household energy consumption in relation to major variables These variables include types of energy using equipment measures of fuel abundance or scarcity and whether fuels are gathered or purchased etc The surveys have varied objectives and differ greatly in the quality and range of data collected and analyzed Nevertheless they can be an invaluable resource for energy assessments

When examined in relation to each other these surveys provide a considerable body of information which can be used to improve the design of future surveys Recent publications have begun to compare and analyze the experiences and methods used in the various energy surveys These comparative publications are very useful reference sources for designing new surveys and interpreting their results (eg Howes 1985)

Local Micro Surveys

Much of the good quality data on household energy use in developing countries has come from small-scale micro surveys These usually cover a maximum of 300-500 households in 10-20 villages but may only cover 5-10 households over a few days Within a limited budget the relatively small samples allow careful quantitative measurements of consumption and related factors although this is not always the case One particularly valuable feature of these surveys is their coverage of qualitative variables such as attitudes to exjsting energy-related problems Indeed the main objective of these surveys often is to understand the social anthropological and micro-economic complexities of household energy demand and supply

Valuable information and insights can also be gained from micro village or urban studies by social scientists anthropologists sociologists argicultural economists and the like These studies do not focus on energy exclusively but nevertheless contain a lot of information on demand and supply and critical linkages in the system For example linkages between the fuel resources system and the total biomass system of village economies may be revealed as well as linkages between the labor and ather demands of fuel collection and cooking and other household activities Any planner working in these areas should always attempt to find these studies

sources Although local surveys and studies can be rich and reliable

of information they generally suffer from four problems

a The quality of data is not always good Fuel consumption in particular often is recorded in terms of weights without any record of moisture content or measured heating values Conversions to energy quantities therefore must be fairly rough and ready

- 32 shy

b Most surveys focus only on fuel consumption and ignore critical supply factors such as local stocks of trees or flows of crop residues which may be the most important determinants of consumption levels and the mix of fuels employed Crucial questions of access to--and hence the availability of-shydifferent forms of fuel by various socio-economic classes (eg the landless non-farm laborers small medium and large farmers) often also are ignored

c Surveys of the same locality at different points in time are extremely rare Consequently they provide little or no information on changes in energy consumption patterns through time or how one group of people responds to trends such as rising income or increasing biomass scarcity

d Good micro-surveys are too few in number to provide an accurate national or sub-regional picture of demand and supply patterns Instead they tend to highlight the enormous diversity in energy consumption An obvious consequence of this fact is that local micro-surveys should never be used as the basis for macro-level assessments or national planning unless there are excellent grounds for thinking that the sample locations are typical or one is content to use rough order of magnitude figures to explore some issue

The force of this last point is illustrated in Table 21 which shows the average per capita consumption of biomass fuels in Ethiopia The figures were estimated in 1980 by the Beijer Institute and in 1983 by a World Bank mission although neither source was based on measured (Le weighed input) surveys The varying results obtained by the Beijer Institute and the World Bank suggest that estimates of national per capita fuel consumption can be inaccurate Also shown are data from towns and cities in very different physical settings based on a third set of measured surveys by the Italian institute CESEN It used quantitative estimates of supply to the whole community though these estimates were not weighed by household consumption

The enormous differences in the regional figures underline the point which cannot be repeated too often that household energy demand and supply must wherever possible be considered at the local level

Table 21 Estimates of Average Per Capita Biomass Fuel Consumption In Ethiopia

(kgyear)

Fuel National Averages

Beijer World Bank Local Data (CESEN) b~ Region Oebre Markos Chefe Moyale

Firewood 424 476 352 1618 417 Dung Agricultural residues

373 232

246 161

77 87

0 3

0 0

(charcoal not shown due to differences in basis of estimates) Sources Anon 11981bl UNDPWorld Bank 11984bl Bernardini 119831

- 33 shy

The paucity of micro surveys and the lack of repeated surveys over time are perhaps the most severe constraints to obtaining a good understanding of household energy demand and supply in developing countries These constraints also limit our understanding of consumers perceptions of their problems and willingness to respond to them as well as the transformations that will occur in the future as conditions change

C MAJOR CONSUMPTION VARIABLES

Several attempts have been made to estimate national average household energy consumption levels by pooling the results of micro and other household surveys A notable exercise of this kind was conducted by FAO for rural households based on nearly 350 surveys and rough estimates in 88 countries [de Hontalembert and Clement 1983] Table 22 shows the results of the exercise

An indication of the iange or local consumption level~ is provided in Table 23 where annualmiddot per capita energy use h shown to vary by a factor of roughly 26 from 23 to 592 GJ or from about 150 to 3800 kg of woodfuel Again the data are for rural areas and are based on national budget surveys or micro surveys in which consumption was measured Table 24 gives comparable data for urban areas

A study of more than 100 household energy surveys shOws that energy use and the choice of fuels consumed depend on mostorall of the following interrelated variables

Supply variables

o Price and availability (for marketed fuels)

o Less easily defined measures of abundance or scarcity especially the time and effort devoted to fuel gathering and fuel use access to fuels by different groups seasonal variation in supply and cultural and socio-economic factors such as gender differences over decision-making and divisions of labor

o The availability of and competition between substitutes for fuel and non-fuel uses of biomass (eg animal fodder construction materials timber for sale small wood for tools etc and soil conditioners or fertilizers)

o Fuel preferences (between biofuels and biofuels versus modern fuels)

o Urban peri-urban or rural location (ie settlement size and proximity to large towns or cities) These differences are closely related to supply factors such as fuel availability

- 34 shy

Demand variables

o Household income

o Household size

o Temperature and precipitation (for space heating and drying needsgt

o Cultural factors (diet cooking and lighting habits number of meals feasts and burial rituals)

o Cost and performance of end-use equipment

Table 22 Annual Per Capita Consumption of Rural Household Energy and Woodfuels Country and Regional Averages and Ranges

Per Capita BiomSS Consumption m Total Pereentage

RegionFuel Type Wood Equivalent GJ as SiCIlIas

AfriCa South of Sahara Lowlands dry 10-15 10 - 14 95 - 98

humid 12 - 15 12 - 14 95 - 98 Uplands (1500m) 14-19 14 - 18 90 - 95 North Afrlea ampMiddle East Larg consumers 02 - 08 2 - 8 Smlll consumers b Mountain areas pound

005- 01 up to 15

05 - 1 up to 15

Asia Including Far East oesert ampsub-desert 01 - 05 1 - 5 Agfleuttural regions dry troples wood fuelS 20 - 50 erop rsldues 02 - 075 2 - 75 20 - 40 animal wastes 045middot 010 4 - 25 20-50 total 065- 105 6 - 10 80-90

Agricultural regions moist tropics wood fuels 20-50 erop residues 03 - 09 3 - 9 20-40 animal wastes 055 - 04 5 - 3 20-40 total 085 - 11 8 - 12 80-90 Shifting agriculture moist tropics 09 - 135 10 - 14 SO-90 Mountain areas wood fuels 125 - 18 13 - 18 6S - 85 other 055 - 02 4 - 2 10 - 25 total f8 - 21 11-20 90 - 95 Latin America hot areas 055 - 090 10 - 14 50-60 temperate areas 070 - 12 12 - 11 55 - 65 cold areas 095 - 16 f8 - 23 50 - 65

Tunisia Iraq Morocco Algeria Turkey bl lebanon Egypt Jordan Syria S ampN Yenene North Africa Iraq Turkey

- 35 shy

Table 23 Per Capita Rural Consumption of Household Energy and Biomass (GJ) Local Averages and Ranges

Average Range Countrysurvey GJ Biomass Source

Bangladesh U I I pur vIII age 68 100 Briscoe 1979 Sakoa vi I 1age 89 70 - 193 97 - 98 Quader ampOmar 1982 4 vi I 1ages 83 large survey 53 95 Mahmud amp Islam 1982 large survey 49 38 - 55 97 - 100 Douglas 1981 budget survey (occupation) 51 37 -61 79 - 91 Parikh 1982

CIIlle 8 vi II ages 292 178 - 592 ( 100) Dlaz ampdel Valle 1984

India large survey (income) 46 43 - 56 92 - 95 Natarajan 1985 Tamil Nadu 4 villages 76 58 - 88 97 - 99 Alyasamy 1982 Tamil Nadu 17 villages 72 42 - 101 97 - 99 SFMAB 1982 Pondicherry (income) 110 102 - 112 91 - 97 Gupta amp Rao 1980 Karnataka 6 vii Iages 10 I 89 - 114 97 - 98 Reddy et al 1980 3 villages 302 76 - 448 96 - 99 Bowonder amp

Ravshankar 1984 Indonesia

3 villages (and Income) 76 53 - 106 45 - 97 Weatherly 1980 Mexico

3 zones (and income) 87 76 - 115 84 - 93 Guzman 1982 Nepal

Pangma v I 1 I age 90 40 - 378 (100) Bajracharya 1981 Pakistan

budget survey (income) 45 35 - 58 81 - 92 FBS 1983 Papua New Guinea

highland village (Jan) 58 25 - 92 ( 100) Newcombe 1984a (May) 54 24- 161 (100) II

South Africa 7 villages 82 52 - 145 ( 100) Furness 1981

Sri Lanka 6 regional zones 84 75 - 112 89 - 93 Wljeslnghe 1984 budget survey (income) 44 23 - 54 86 - 92 DCS 1983

Tanzania 18 vi I I ages 109 44-261 ( 100) Skutsch 1984

Note Ranges are not for Individual households ranges for them are much greater These ranges apply to averages at one level of disaggregation below the average shown in the table eg income or caste groups in a one-village survey

- 36 shy

Table 24 Per Capita Urban Consumption of Household Energy and Biomass (GJ) Local Averages and Ranges

Countrysurvey

Bangladesh budget survey (occupation)

India Hyderabad (Income) a I arge survey (i ncome) Pondicherry (Income)

Pakistan budget survey ( income)

Papua New Guinea squatters settl~nts government housing

settlements high income housing

Sri Lanka budget survey (income)

Togo LOIIe (income)

Average

35

24 33 59

30

11 2

83 236

30

51

Rllnge GJ

34 - 35

21 - 29 31 - 39 57 - 66

27 - 48

135 - 337

23 - 38

46 - 55

bull Biomass

49 - 67

26 - 72 36 - 78 70 - 84

25 - 80

79

41 lt1

22 - 87

Source

Parikh [19821

Alam et al (1983) NataraJan (1985) Gupta ampRao [19801

FBS (1983)

Newcombe [1980)

DeS (1983)

Grut [19711

a Excludes electricity use b Wood fuels only Note Rangesmiddot are not for Individual households those ranges are much greater These

ranges apply to the averages at one level of disaggregation below the average shown in the table eg income or caste groups in a one-city survey cities or towns in a multi-ciTY survey an~ income groups in a natlonjll urban survey

The main effects of these variables are examined below At the outset i~ should be obvious that many of them overlap and that there is often no clear distinction between variables that affect demand and supply For example the cost of end-use equipment is listed as a demand variable since it concerns the final end of the energy supply-conversion chain and is linked to factors such as income preferences for using certain fuel s and even tastes in the case of cooking equipment But end-use technologies are often fuel-specific as with a kerosene lamp or stove and so depend on supply-side issues stich as the availability and price of fuels and the price of household equipment Some other factors which are known to have major effects on consumption in developed country households including dwelling size and daily occupancy patterns are not listed because there is virtually no information on their effects in developing countries

- 37 shy

Gathered Fuels and Time Budgets

A fundamental division is made between households which gather fuels and those which buy them This distinction is not always clearshycut since fuel gatherers may hire a donkey or truck to collect fuel from a distant source or pay for fuels by bartering goods services or their own labor Many gatherers also buy some modern fuels such as a little kerosene for lighting or for starting the cooking fire and many households gather or buy traditional fuels at different times of the year

Nevertheless the distinction 18 an important one for two reasons

a It emphasizes the contrast between local and macroeconomic issues Fuel gatherers have access only to local resources Buyers are part of a more generalized national system of prices and energy delivery infrastructures

b Gatherers pay for fuels by complex trade-offs between fuel preferences fuel economies and time available for energyshyrelated and other household or productive activities Their access to fuels is often governed by local rules on rights to use common land and client-patron relationships concerning the land of neighbors Buyers tend to respond to conventional market forces

For poor families and especially for women in many societies time 1S the major factor of production and a scarce resource [Cecelski 1984 Thus time expenditures on energy-related tasks are a major factor in household decisions about the level of energy consumption and the types of fuels used

This decision process which is not simple has been well summarized by Cece1ski [1984

Rural households make decisions on the relative values of time in cooking and labor of household members during different periods versus the cost and convenience of alternative fuels Most of these decisions are made by women but women do not always control income spent on fuel or the fuel types selected by other family members Interactions within the household determine a total systems efficiency of fuel procurement and use to optimize labor and cost Seasonal agricultural peaks can intensify labor and fuel demand conflicts

Table 25 indicates the range of fuel collection times that have been found in surveys in person-hours per household they range from 8 minutes to 38 hours per week However other fuel-related time factors must also be considered including fuel preparation (eg wood cutting and splitting breaking and bundling crop residues making dung cakes)

- 38 shy

procuring alternatives such as kerosene food preparation and cooking and fire tending All these factors must be judged alongside other time demands as well as alternative uses of biomass such as house construction material thatching animal feed and fertilizer

Table 25 Fuelwood Collection Times (Hours per Week per Average Household)

Country VI I 1 age Mean Range Source

Bangladesh (1 v I II age) 25 White (9761

Burkina Faso rural 09 McSweeney (1980 )

Chi Ie (7 vi I I ages) 118 50 - 255 Diaz amp del Valle (1984)

India Karnataka (6 viii) 116 84 - 164 Reddy et al [19801

T Nadu (4 viII) 95 26 - 186 Alyasamy et al 119821

Indonesia Java 21 White (1976)

Long Segar 014 Smith amp Last 11984 )

Kal I Loro 063 Smith amp Last [1984)

Nepal (6 v I ages) 43 Acharya amp Bennett ( 19811

(1 vi II age) 22 94 - 38 Spears [1978)

Peru (3 v i II ages) 35 - 116 Skar [1982 )

S Africa (3 v I II ages) 113 - 148 Best 11979)

Tanzania (18 vi I I ages) 93 12 - 212 Skutsch 11984)

Lushoto 10 - 18 Fleuret amp Fleuret (1977)

Due to these complexities the relationship between physical measures of fuel scarcity and how people perceive the costs of fuel gathering is rarely simple Although as a general rule greater fuel scarcity equates to greater collection distance and time and hence to fuel substitutions and economies these generalizations should always be checked Local exceptions to the rule may spell failure for any project which is based on common expectations Some examples of exceptions and key points to watch out for are given below

- 39 shy

Strong fuel preferences frequently override time considerations For example in one Tanzanian Maasai village women walked several kilometers to chop wood from a particular species of living tree returning with backloads of up to 60 kg even though the nearby forest floor was littered with fallen branches of other wood species The more distant species could be lit without any kindling wood or kerosene and burned for a long time with a steady flame [Leach 1985] A large survey in Thailand found that distance to the fuel source and collection time had no impact on consumption levels or the replacement of wood by other fuels In this case there was a strong tradition of using wood as opposed to charcoal or kerosene [Arnold amp deLucia 19821

Seasonal factors may be important In particular the demand for labor in peak agricultural seasons often imposes severe time conflicts and leads to temporary reductions in fuel gathering and consumption In Pangma village Nepal the average wood collection trip took 5 hours to gather a 40 kg bundle In the peak agricultural season this was considered a burden But in the slack season going to the jungle for wood was a chance for a group outing and singing dancing gossiping and joking Substantial differences in consumption were noted due to seasonal rather than other factors [Bajracharya 1981]

Collection time may not be related to distance in which case it is almost invariably time and not distance that is the key factor This could happen when the nearest wood resources are at the top of a steep hill for example as in one area of Lesotho [Best 1979] Scavenging low quality fuels near the home may take longer than getting firewood from a more distant source but may still be preferred because small amounts of fuel can be gathered rapidly This collection pattern was frequently observed in the large Malawi rural energy survey [French 1981] for example among women who were caring for young children and could not leave home for long periods

Fuel economies are often judged according to complex time considerations Although it might seem obvious that saving fuel would save time on fuel gathering economy measures may also consume considerable amounts of scarce time -- for example the careful tending of the cooking fire Energy savings therefore depend on a woman s complete time budget [Koenig 1984] One consequence is that saving time in cooking is often given a higher priority than saving fuel so that the cooking methods employed use more fuel than they would if time were not limited In Tanzania [Ishengoma 1982] and Senegal [Madon 1982] women were interested in improved stove designs mostly because they saved cooking time rather than cooking fuel

Time constraints are often greatest for the poorest When fuels are very scarce women are often forced to work even longer hours than usual or get other family members--usually children--to take over some of their workload These adjustments are obviously more difficult in small households or where an adult member of the family is old sick

- 40 shy

or disabled conditions often associated with extreme poverty For example a survey in Orissa India found that half of the families had seriously reduced the time spent on household tasks in order to collect sufficient fuel and that the consequences were most damaging in families which were both the smallest and the poorest [Samantha 1982]

Buying fuels is often the last resort for poor families However when the decision is made to purchase fuels it frequently is based on time considerations Trade-offs are made between (1) the costs of fuels and the equipment to use them and (2) travel times and costs to reach fuel markets time saved in fuel gathering and the opportunities to earn cash in the time saved

Time Costs of Fuel Collection

The previous section emphasized the critical importance of time constraints for fuel gatherers A useful way of assessing and comparing these costs is to estimate the rate of fuel collection and convert it into a monetary value to give a cash measure of the opportunity cost of fuel collection

An example of such a calculation based on a Mexican village [Evans 1984] shows that the opportunity cost of firewood collection may be very high The average collection rate was 62 kghour while the local market price of wood was MN$ 3 per kg The value of wood collecting was thus MN$ 186 per hour The minimum laboring wage at the time was MN$ 275 per hour If jobs were available it would be more cost effective to earn cash as a laborer in order to buy wood than to collect it

The fuel collection rate is also valuable as a single measure of fuel scarcity It combines in one figure most of the pertinent information provided by other commonly used indicators such as distance to fuel sources collection time and density of the fuel stock at the collection site and it does so for the two quantities that matter most to families fuel consumed and the time cost of gathering it

Table 26 shows the wide variation in collection rates For average conditions in these surveyed locations the range is from 17 kghour in South India to more than 70 kghour in the Chilean subsistence village close to forest resources In all these cases wood was collected on foot and by headload or back10ad Where animals (or trucks) are used rates may of course be higher for the same conditions of fuel scarcity

- 41 shy

Table 26 Collection Rates for Firewood (kghour)

Country V I I I age Mean Range Source

Chile (7 villages) 265 125 - 714 Diaz ampdel Valle (1984] India Karnataka (6 villages) 28 17 - 38 Reddyet al (1980]

Tamil Nadu (4 villages) 39 18 - 54 Aiyasamy et al (1982) Indonesia (3 vii 1ages) 10 - 20 Weatherly (1980] Mexico (2 villages) 62 - 92 Evans (1984] S Africa (3 vii Iages) 55 38 - 67 Best 1979] Tanzania (18 villages) 121 43 - 444 Skutsch (1984] Yemen (8 villages) 36 Au Iaq i (1982]

Income and Rural-Urban Differences

Income and rural-urban location are among the strongest variables in determining total household energy use the mix of fuels employed and consumption for the major end-uses such as cooking lighting and electrical appliances They are best considered together as income has different impacts on fuel consumption patterns in rural and urban areas

The broad effects of these variables on energy use can be seen in Figures 21 and 22 which are based on large nationally representative surveys for Brazil (1979) India (1979) Pakistan (1979) and Sri Lanka (l982) [Goldemberg 1984 Natarajan 1985 FBS 1983 CBC 19851 Several points are immediately obvious

Energy consumption is much lower in urban than rural areas especially for middle income groups This is mainly because these groups in urban areas can obtain and afford high efficiency modern fuels and equipment to use them On a utilized energy basis the ruralshyurban differences would not be so great Figure 22 confirms this point by showing the share of traditional biofuels in total energy use across household income In rural areas there is virtually no change with income and the shares are all within 85-95 the remainder being mostly kerosene for lighting In urban areas the lowest income groups also depend mostly on traditional fuels with shares close to 80 except for Sri Lanka (90) As incomes increase the share of traditional fuels drops sharply to a minimum of around 25-30 again except in Sri Lanka The substitution of modern for traditional fuels in these cases depends on (a) urbanization and (b) rising urban incomes

bullbull bull

bull bull bull

- 42 shy

FIGURE 21 Household Energy Consumption against Household Income Rural and Urban Areas in Brazil (1979) India (1979) Pakistan (1979)

and Sri Lanka (1982)

Rural so

Srazil

India bullbullbullbullbullbullbull bullbullbullbull Sri Lanka

~

J bull bull

bullbullbull bull Pakistan

bull I

I bull

bull

~ I (

I

OL--L~__L--L~__~~~__~~~__~~~__~~~~

o 2 4 6 S 10 12 14 16 is Household Income Thousand USS (1975middot PPP Corrected)

Urban 40 bullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbullbull Sri Lanka

bull

bull _ Pakistan

India

bull -- - - ~r- _~ ~ ~ ------------------------------~B~ra~zil

bullbull

Household Income Thousand USS (1975 PPP Corrected)

Note bull PPP =Purchasing Power Parity World Bonk-307361

- 43 shy

FIGURE 22 Share of Biomass Fuels in Total Energy against Household Income Rural and Urban Areas In Brazil (1979) Pakistan (1979)

and Sri lanka (1982)

Rurol 100

Indio

~WlI4~~Jfr~middot~-imiddot~~middot~~~~middotmiddotmiddotmiddot~middotmiddot~middotmiddotmiddot~middotmiddot~middot~middot~~sn~middot~Lon~ko~____ Brazil

Pakistan CD

805s () gtshy ~ w

QZ J

~ in

o 2 4 6 8 10 12 14 16 18

Household Income Thousand USS (1975 bull PPP Corrected)2

Urban

bullbullbull bullbullbullbullbullbullbull bullbullbullbullbullbullbullbullbullbullbullbullbull Sri Lanke

80

gtshy~ c w ~ 0- J 40

I India

bull _ bull _ bull _ bull 2kstan ----=~------ Brazil

20

o 2 4 6 8 10 12 14 16 18

HousehOld Income Thousand USS (1975 bull PPP Corrected)

Noles bull inclUdes energy consumption by hOusehOld members and servant 2 PPP Purchasing Power Porlty

World Bonk-307362

- 44 shy

The exceptional behaviour of sri Lankan urban households is explained by another major variable fuel availability (and prices) In urban Sri Lanka as much as 30 of domestic firewood comes from the households own lands or garden compared to an average of 25 in India and when firewood is purchased its price at the time of these surveys was close to 60 of that in urban Pakistan and 40 of that in urban India 2

One also sees a strong and fairly steady relationship between total energy consumption and income and a marked tendency for energy use to rise steeply at low incomes but to saturate at high incomes Discussion of these trends is deferred to the next section on the effects of household size

Although these trends are useful general indicators they are less important to understanding household energy use than are their underlying causes Five of these can be singled out as they are found in many countries and explain much of the variation in fuel mix among income groups total ener~y and rural-urban locations

With increasing income one normally sees

a Steady or increasing biomass consumption in declining biomass consumption in urban areas

rural areas but

The rural trend is explained by easier access to biofuels since land or cattle ownership is greater and by the ability to purchase biofuels The urban trend is explained) by the fuel substitutions described below and by the tendency to eat more meals outside the home thus reducing cooking needs

b Substitutions between urban areas

biomass fuels for cooking especially in

For example in urban Africa and Latin America charcoal often displaces firewood as the main cooking fuel This is partly a matter of taste but also of convenience charcoal is easier to transport and store and less smokey than firewood The degree of substitution and the income level at which substitution begins depend on the relative prices of firewood andmiddot charcoal and the relative costs of cooking equipment as well as cultural preferences

c Substitutions of modern especially in urban areas

fuels for biomass cooking fuels

pound Prices compared between countries by normalizing to the US$ with Purchasing Power Parity indices [Leach 1986]

- 45 shy

With increasing income the progression is normally kerosene - gas (eg LPG) or electricity

biofuels -

d Greater use of modern fuels and electricity for end-uses than cooking

other

With lighting typically there is an increase in kerosene use followed by a decline at higher incomes as electric lighting is installed This trend is usually strongest in urban areas where kerosene and electricity are more widely available and depends on equipment costs as well as relative prices The other major trend is a rapid expansion of electricity use for refrigeration space cooling and other electrical appliances This typically begins at low to middle income groups in urban areas but only at high income levels in rural areas (although this depends on the extent of rural electrification the cost of hook-ups to the grid and the price of electrici ty) bull

e A tendency for consumption of modern fuels highest income levels

to saturate at the

In many developing countries without significant space heating needs energy consumption by urban households at the highest income levels clusters around 25-35 GJ per family per year This is close to 20-25 of household consumption at equivalent incomes in industrial countries or much the same as the industrial country level when space heating is deducted

increases shortages

These trends reflect two underlying forces As spending power in rural areas families can buy their way out of biomass fuel andor have sufficient land to grow their own biofuels In

both rural and urban areas greater purchasing power pulls families toward more efficient and convenient modern fuels and the new end-uses they allow Except at the highest incomes when space cooling is introduced there are marked limits to the amount of energy required to satisfy these end-use needs (eg lighting refrigeration and other electrical appliances)

The progression from using biomass fuels for cooking to using kerosene LPG and electricity as urban incomes rise is shown in Table 27 The large differences between the cities are due to differences in average income degree of modernization and energy supply infrastructures

Household Size

With nearly every household use of energy there are large economies of scale associated with increasing household size For example the additional energy required to cook for four persons rather than two is small compared to the fixed overheads for keeping the fire

- 46 shy

alight etc With lighting and space heating energy use depends on the dwelling area or number of rooms other things being equal and is not much greater for a family of four than for a family of two

Table 27 Cooking Fuels Used In Urban Households (percent of households In fuel grouping)

CltylHousehold Type Firewood Charcoal Kerosene LPG Electricity

Kuala Lumpur (1980) Low income 4 15 75 25 19 Middle income 7 23 57 52 35 High income o 17 19 87 50

Mani la (1979) Low income 9 35 45 11 Middle income 2 1 5 73 19 High Income o 78 19

Hyderabad (1982) Low income 41 (a) 70 19 (b)

Middle income 24 (b) 65 54 (b)

High income 13 (b) 57 71 (b)

Bombay (1972) Low Income 10-30 10-30 98 9 Mi dd Ie income 3-20 3-20 98 53 High income 3-10 3-10 77 94

Papua New Guinea (1978) Low Income 79 21 Middle income 41 42 17 High Income 0-6 0-7 87 - 93

Note Data for Kuala Lumpur and Hyderabad reflect use of more than one fuel Man I I a data refer to usua I source of energy Bombay data refer to ownership of cooking devices The percent of Bombay households owning a hearth for burning firewood or a stove for burning coal was 40 23 and 13 for the respective income groups (a) Sma I I amounts of charcoal are used at all income levels (b) Not measured

Sources Sathaye ampMeyers [19851 based on SERU (1981) (Kuala Lumpur) PME [19821 (Manila) Alam et al [19831 (Hyderabad) Hernandez (1980) (Bombay) Newcombe 119801 (Papua New Guinea)

This effect is illustrated schematically in Figure 23 In the left-hand figure total energy consumption rises linearly with household size so that per capita consumption falls steeply at first and then flattens out In the right-hand figure total energy rises rapidly at first and then grows more slowly so that per capita consumption remains roughly constant

-----

---------

- 47 shy

FIGURE 23 Effects of Household Size on Total and Per Capita Energy Consumption

Household size often has as great or greater an effect on energy consumption as other major variables such as income Furthermore in some countries household size is strongly associated with income on average large families tend to have more income earners while high income households may attract family relatives This is certainly the case in South Asia Consequently when the data shown in Figure 21 is replotted for the South Asian countries on a per capita basis (see Figure 24) there is little variation in per capita energy consumption across the entire household income range In other words the rising curves for household energy plotted against household income (Figure 21) are mostly a function of increasing family size with household income

These effects are of great importance when comparing and assessing survey data or using them to project energy consumption First whenever absolute levels of consumption are important (as opposed to fuel shares etc) it is obvious that one must work either in per household or per capita terms But since many surveys do not publish data on household size which allow conversion between these bases the range of surveys that one can use may be limited Note though that the survey authors may be able to provide the missing information on household size

f

Total

1

_ Per Capita

Household Size --

f ~ ltJ)c w

Total

Per Capita

Household Size ---t

World Bank-307363

bull bullbull

- 48 shy

FIGURE 24 Per Capita Energy Consumption against Household Income Rural and Urban Areas in Pakistan (1979) India (1979)

Rural

bull 10 -

Sri Lanka bull8

( Q)

~ (] gt 6 Indio

~ c bull

- - - bull __---shy Pakistan

1bull~ -_ shyw _-shy __ ~ 0 0 4 U (j) 0

2

o 2 4 6 8 10 12 14 16 18

Household Income Thousand USS (1975 - PPP Corrected)

Urban

8 Sri Lanka0 bullbullbullbullbullbullbull Q)

~ bullbullbullbullbullbullbullbullbullbull ltD e

gt 60gt ee

(j) c w

Ea bull India u ~ - ---__ __-Pakistan 0

--r ----shy~ ---__-_ - 2

O~~~__~~~__L-~~__L-~~__L-~~__L-~~~

o 2 4 6 8 10 12 14 16 18

PPP = Purchasing Power Porily

World Bank-3073611

- 49 shy

Second whether per capita or household energy data are used one has to be wary of the effects of household size This warning applies particularly to the use of regression methods to estimate energy income elasticities A formal description of this problem is given in Table 28

Third it is usually sufficient to base assessments on per capita data (the kind most frequently reported) and to combine these with total population and its growth rate to derive total consumption However if there is any cause to believe that household size is likely to change appreciably (eg for different income groups) then projections of household formation rates andor average household size will also be needed

Table 28 Relationships between Energy Income and Household Size

Household energy frequently depends closely on household income according to a relationship such as

o = a yb ( denotes multiplication) where (0) is the consumption of a fuel or total energy (y) is household income (a) is a constant and (b) is the energy-income elasticity Regressions of survey data using this equation often show that income explains at least 90-95 (or more) of the variance in energy use However energy use also depends strongly on household size whi Ie household size may be

closely linked to household income In other words N =c yd

and 0 = e Nf

where (N) is household size (c) and (e) are constants and Cd) and (0 are elasticities If these expressions are combined and manipulated it can be shown that (i) there is no simple expression linking per capita energy and per capita income and (ii) that the only simple (two term) relationship is the one linking per capita energy and household income It is for this reason that In Figure 24 per capita energy is plotted against household income rather than say per capita income The four most obvious and useful relationships are shown below

1 Household energy to Household Income and Household Size b-do = alc y N

2 Per Capita Energy to Per Capita Income and Household Size (QN) = a (YIN)b Nb-

3 Per Capita Energy to Per Capita Income and Household Income (ON) =a cb- 1 (YN)b yd(b-l)

4 Per Capita Energy to Household Income (OIN) = alc yb-d

- 50 shy

Purchased Fuels and Expenditure Shares

The share of income or expenditure devoted to providing energy is an important factor in assessing household fuel use If the share is very high it indicates that families are severely stressed by their energy problems and are likely to welcome solutions If the share is low families may be indifferent to rising energy prices or increased fuelwood scarcity as well as attempts to introduce energy saving measures

In both developed and developing countries the lowest income groups spend the largest shares of their incomes on energy This point is demonstrated in Table 29 for urban households where most fuels are purchased Data for the US and UK in the early 1980s are included for comparison

Table 29 Household Budget Shares for Energy in Urban Areas (percent)

Lowest Highest Mean Income Income Source

USA 1982 01 I heatl ng 82 319 36 EIA 11983] aII househol ds 45 200 27 EIA ( 1983]

UK 1982 62 119 43 DOE ( 1983)

Brazi I 1979 190 09 Goldemberg et al (1984)

Chi Ie (Santiago) 1978 42 76 31 Anon [19831 1968 41 47 33 ILO (1979)

Egypt 1975 36 46 30 ILO ( 1979)

India Hyderabad 1981 al 36 107 15 Alam et al [ 1983) Pondicherry 1979 184 52 Gupta amp Rao ( 1980)

Lesotho 1973 48 88 37 ILO [ 1979)

Pakistan 1979 40 86 18 FBS [ 1983)

Panama 1980 20 Anon (1981a)

Sri Lanka 1981 47 97 32 DCS [19831

Excluding electricity

Note Budget shares for energy are def I ned as the percentage of income or expend i ture devoted to househo I d f ue Isand e I ectr i city exc I ud I ng motor veh i c 1 e fue Is Non-marketed gathered f ue 1 s are I nc I uded us i ng an imputed price In urban regions this probably has I ittle effect on actual cash expenditures on fuels

- 51 shy

Even higher budget shares than those shown in Table 29 are often cited for particular cities or regions of developing countries Examples are 20-30 in Ougadougou Burkina Faso [Anon 1976] 30 in the town of Waterloo Sierra Leone [Cline-Cole 1981] and 25-40 in the capitals of the Sahel region of Africa [Lambert 1984 Wherever the original sources for such widely-quoted figures can be tracked down it usually turns out that they refer to special groups such as low incomeshyearners with large families or even a single household with an unusually high share of income devoted to energy costs Such figures therefore have to be used with considerable caution when considering the effects of prices or incentives to reduce expenditures through fuel saving measures etc for all income groups or the whole population

Energy Prices

Many attempts have been made to use differences in energy prices to explain variations in consumption levels and fuel choices in different countries Unfortunately this approach is severely hampered both by the lack of reliable data on local energy prices and also by the problem of converting prices to a standard unit such as the US dollar To reflect true differences across countries prices should be converted to US dollars using purchasing power equivalent exchange rates In low income countries these increase the real equivalent dollar price of goods and services by a factor countries by around 15 to 3 times

of 3 to 35 and [Kravis 1982] 11

in middle income

Alternative approaches are to compare countries using (1) shadow exchange rates or (2) an index such as price relative to average per capita income Table 210 presents estimates of fue1wood and charcoal prices and average daily wages for several countries As a percentage of average daily wages prices vary from less than 1 to more than 13

Table 210 Relative Prices of Woodfuels in Selected Countries

Market Market Average Price Price Percent

Dai Iy of 15 KG of 05 Kg of Daill Minimum Wage Country Wage Firewood y Charcoal Firewood Charcoal

Ethiopia 200 Birr 021 Birr 022 Birr 135 110 Madagascar 100000 FMG 3300 FMG 2150 FMG 33 28 Malawi 100 Kw 006 Kw 008 Kw 60 80 Sudan 200 SL 008 SL 008 SL 42 39 Zambia 364 Kw 003 Kw 006 Kw 08 16

al Solid wood stick bundles Source World Bank Mission staff measurements and observations

31 This reference provides equivalent (or parity) exchange rates for a number of countries

- 52 shy

Within a given country the usual methodmiddot of determining the effects of prices on consumption and fuel substitution is to estimate the price elasticity of demand (see Chapter I Section D) This estimate normally differs depending if income is constant or changing so the income elasticity of demand must also be estimated Both estimates require time series data on consumption income and prices Furthermore data for many years is required to distinguish immediate reactions to higher prices from the more stable and usually much smaller responses over the longer term As discussed before this information is rarely available for the household sector in developing countries

As a result in most developing countries there is remarkably little information from which to judge how even at the most aggregate level households will respond in their fuel consumption to changes in income or fuel and power prices Other methods of projecting energy demand particularly for biomass fuels are reviewed in Chapter V which also discusses the roles of fuel pr1ces in assessing alternative technologies such as cooking stoves

D ADAPTATIONS TO FUEL SCARCITY

A useful perspective on consumption differences can be gained by considering the responses that people make to the depletion of woodfuels the major household energy source in developing countries

Adaptations in Rural Areas

As a starting point in some rural areas abundant fuel grows virtually on the doorstep Fuel collection is a relatively trivial task Consumption is unconstrained often abnormally high (especially in colder areas) and only preferred species of wood are used This may be true even in areas within countries where biofue1 supplies are generally scarce

Under these conditions an annual fue1wood consumption of up to 4 tons per person has been estimated for subsistence communities living close to the forest in the colder regions of Chile 41 Annual consumption levels of 29 and 26 tons woodfue1 per person have been reported for fairly high altitude areas of Nicaragua and Tanzania respectively [Jones amp Otarola 1981 Fleuret amp F1euret 1978] In warmer regions where demand is mostly restricted to cooking and water heating unconstrained consumption levels seem to fall in the range of 12 - 15 tons per person per year

41 This level of consumption is estimated from the following formula based on Table 23 60 GJ x 1000 t = 4 tonnes

15 GJ

- 53 shy

For the majority of rural households fuel collection is more difficult and has appreciable personal costs in terms of time and effort With increasing scarcity one generally finds the following broad stages in adaptation

a Lower quality but more accessible woodfue1s are used This expands the resource base and may postpone the need for any further adaptations Where population densities are low demand can often be met without depleting the standing stock of trees Families who own sufficient land are often able to meet their demand from their own resources others can usually collect from nearby forests common lands roadsides or wastelands

b People start to economize on fuel This normally occurs when the time required to collect wood has become an unacceptable burden For example cooking fires are smaller embers are quenched after cooking for re-use later or greater care is taken to shelter the fire from the wind Some least essential end-uses such as water heating for bathing or washing clothes and dishes may be reduced Consumption drops considerably Typical figures are hard to define but from the evidence of many surveys in areas without significant space heating consumption appears to be in the range of 350-800 kg per person per year This level of adaptation may coincide with the first signs of interest in fuel-saving stoves

c Crop residues and animal wastes begin to be used This adaptation is found right across the developing world and is often seen as an easier (ie less time consuming) response than tree planting The adaptation may be most difficult for the poor andor landless who must depend on supplies from other peoples land and animals or common land As biomass supplies of all kinds are depleted traditional rights of access to fuel sources are often closed off to the poor

d Reductions in living standards and diet are found in conditions of acute scarcity Income-earning tasks hygiene child feeding and care or visits to health and education services may be reduced or e1 iminated in order to make time for fuel gathering [Cece1ski 1984] Fuel and hence time may be saved by reducing the amount and kinds of cooked foods in the diet Staple foods which require less cooking are introduced food may be re-heated rather than cooked a fresh processed foods are purchased and the number of meal s may be reduced Some examples ascribed to fuel shortages are greater consumption of raw foods in Nepal [Cecelski 1984] and reductions in staple beans in Guatemala Mexico and Somalia [Tinker 1980 Evans 1984 Cecelski 1984] However it is not always clear that fuel shortages are directly responsible for these or other examples of food deprivation A reduction in dietary

- 54 shy

quantity and quality may reflect an attempt simultaneously to save money time and fuel

e The purchasing of biomass or modern fuel substitutes by people who previously collected them free is another important response to scarcity--not just of fuels but also of fuelshycompeting materials such as animal fodder Essentially the judgment is made that the benefits from alternative uses of biomass fuels (eg straw for fodder rather than fuel) or the time saved from fuel gathering is greater than the financial burden on often severely limited budgets for fuel purchases Since this decision framework is complex while there are large differences in the price and availability of commercialized fuels the degree to which this occurs varies enormously

fuel can emphasize

These adaptations suggest that consumption levels and types of vary greatly in response to deepening fuel scarcity They the dangers of extrapolating present consumption patterns into

a future of greater woodfuel scarcity or of supposing that a shift away from woodfuels to modern fuels will occur automatically as incomes increase as it has in developed countries National energy plans have frequently been rooted firmly in one or the other of these notions

Perhaps most importantly these adaptations underline the critical distinctions between households who own land and those who do not in determining their ability or willingness to plant trees in order to alleviate their fuel shortages Their incentives to do this are not a matter of average supplydemand balances--the fuelwood gapstl that the outsider frequently measures They stem from personal perceptions and balances between present costs of fuel collection and the costs and benefits of many alternatives of which tree planting intended primarily for fuel supply is only one

People who have little or no land often feel the effects of fuel scarcity most acutely but are at the same time least able to respond by planting trees or burning crop residues and animal wastes Those who have land often may have sufficient fuel for their needs or need little help in planting a few trees to provide more fuel If the latter are to be induced to grow more fuel than they need themselves there must be (1) a market in which to sell it and (2) a market which provides a greater return on investment than alternative uses of their land and labor

In many locations in developing countries these market factors are dominated by the demands of urban areas which can extend many hundreds of kilometers into the hinterland (see Chapter III) In these cases urban demands for woodfuels are one of the principal causes of rural woodfuel depletion but also provide the major opportunity for increasing (commercialized) rural fuel production

- 55 shy

In other areas rural traditions of gathering wood without any cash payments are increasingly giving way to commercial wood markets As mentioned above the extent to which rural commercialization of woodfuels has already occurred varies greatly In Tanzania only salaried public servants such as teachers -- or less than 25 of rural families -shygenerally purchase their firewood (Nkonoki 1983] In Malawi 10 of rural families purchase firewood but only 40 of their needs are met in this way (French 1985] In other countries with higher incomes better developed rural infrastructures or greater fuelwood scarcity this process has gone much further In Nicaragua for example some 40 of rural consumers buy some or all of their wood (Van Buren 1984] while in the arid mountainous Ibb region of North Yemen 65 of rural households buy a quarter or more their fuel (Aulaqi 1982)

Adaptations in Urban Areas

For the urban and peri-urban poor gathered or purchased woodfuels are the major energy source Responses to greater scarcity (or higher prices) are much the same as those listed above economies and lowered fuel quality standards People buy or scavenge trashtl fuels such as small wood pieces sawdust and mill wastes etc However for many urban families living in high density apartments or small houses biomass fuels are often ruled out due to lack of space for storage and drying and frequently lack of a chimney or flue for the fire Hence the most prevalent fuels are all commercialized charcoal and modern energy sources such as kerosene bottled gas (LPG) and electricity

Another major class of response for the poor is a price-driven substitution of modern cooking fuels for fuelwood (or other traditional fuels) This almost invariably means kerosene rather than the other major alternatives LPG and electricity Kerosene stoves are relatively cheap and portable (an important factor for shanty dwellers and itinerant laborers who may have to move homes quickly) The price of bottled gas cylinders and gas stoves and of connection to the power grid (assuming this is possible) is normally prohibitive to the poor and lower-middle income families

Urban consumption patterns are also strongly driven by incomeshyrelated substitutions of modern fuels for biofuels Since the former are generally available in large towns and cities as incomes increase families can afford to attain the higher living standards offered by modern cooking fuels such as greater cleanliness convenience and efficiency At the same time families benefit from new end-uses offered by electrification such as better lighting refrigeration and for the highest income groups space cooling Urban energy behavior thus is much more like that of developed countries and depends largely on income the price of energy and the cost of energy-using equipment In developing countries the availability of fuels (especially LPG and electricity) is an important additional factor large cities tend to have a more modernized pattern of fuel consumption than medium or small towns

- 56 shy

because electrici ty and LPG (and piped gas in some countries) are more widely distributed

The strength of these urban substitutions and hence the possibility for rapid changes in energy demand patterns are illustrated in Tables 211 and 212 using data for India [Natarajan 1985 1986]

Table 211 shows the effects of settlement size in India on the fuel mix for cooking and heating In towns with populations of less than 20000 modern fuels provide about 39 of utilized energy for these endshyuses but in cities with more than 500000 residents the share is close to 75 With LPG the share increases tenfold across the urban size range The table provides a sharp reminder that the usual simple division of households into rural and urban may be wholly inadequate urban size as well as the proximity of rural areas to neighboring cities and transport routes may be critical factors because of their effects on the availability of modern fuels

Table 211 Household Energy Patterns and City Size India 1979

City Size (thousand Per Capita Percentage Shares of Modern Fuels a residents) Energy All Electricity Kerosene LPG Coke

OYer - 500 294 754 135 289 156 173 200 - 500 275 662 94 286 130 142 100 - 200 269 575 92 198 72 213 50 - 100 266 562 80 187 64 225 20 - 50 234 376 63 95 29 188

Under 20 244 390 67 166 1 5 143

All 266 570 93 212 85 177

Energy totals and shares are given in terms of kilograms coal replacement an approximation to useful energy Small amounts of town gas are omitted

~ NataraJan [19851

Table 212 shows how very rapid transitions from traditional to modern fuels can occur in urban areas During 1979-84 firewood prices rose quite steeply in most Indian cities while the prices of kerosene and LPG fell in real terms [Leach 1986J During the same short period as shown in the table the share of firewood in cooking and heating dropped from 42 to 27 on a utilized heat basis The shares of kerosene and LPG almost doubled The greatest reductions in firewood use took place in the middle income groups but the poorest households also reduced their shares (from 60 to 535) This table highlights both the possibility for fuel modernization as a solution to increasing

- 57 shy

Table 212 Fuel Shares tor Cooking and Heating by IncOllle India 1979 and 1984 (percentage shares)

------------------Income---------------- shyFuel Type Year L LM M liM H All

Firewood 1979 600 409 251 17 4 121 424 1984 535 308 179 99 96 274

Soft Coke 1979 128 202 236 167 17 3 184 1984 64 180 179 152 83 153

Kerosene 1979 132 213 215 220 189 187 1984 238 369 402 382 328 357

LPG 1979 08 46 142 269 329 66 1984 152 97 83 88 101 101

Other 1979 133 131 156 170 188 139 1984 152 97 83 88 101 101

Percentage 1979 (315) (428) (207) (26) (24) ( 100) of households 1984 (176) (336) (351) (94) (43) ( 100)

Incomes (Thousand Rupees IRs 1978-791 a year) L Low (under 3) LM = Low-middle (3-6) M=Middle (6-12) liM = High-middle (12-18)1 H High (over 18)

Shares are on a coal replacement basis tor cooking and heating

Source Natarajan [19861

scarcities of traditional fuels and the need for developing countries to conduct regular large-scale household energy surveys to track consumption trends over time

E ENERGY END-USES

A households total energy consumption and mix of fuels is the result of the familys attempt to provide for its various needs by employing its labor or cash and specific technologies that use a certain type of energy The micro-perspective of each consumer is therefore the driving force behind the sectors use of energy and opportunities for change in demand and supply patterns In this section we examine briefly the relative importance of the major energy end-uses Chapter III goes

- 58 shy

into them in greater detail and includes discussions on the efficiencies and costs of end-use equipment

Among the poorest families in most developing countries cooking (and heating) accounts for 90-100 of fuel consumption the remainder being for lighting by the cooking fire kerosene lamps candles or electric torches At higher incomes better lighting is one of the first priorities in order to improve living standards and frequently to extend the working day At still higher incomes water heating refrigeration and cooling begin to play an important role The need for space heating may well decline since dwellings are generally better constructed

A classic pattern of this kind can be seen in Table 213 which is based on a large rural survey in Mexico taken in 1975 [Guzman 19821 In each of three regions as incomes rise the shares for cooking decline the shares for water heating increase sharply and the shares for space heating first increase and then decline Energy for lighting is not included

Table 213 End-Use of Energy for Cooking and Heating in Rural Mexico (Percentage Shares)

Zone 1 Income Zone 2 Income Zone 3 Income End Use Low Mad High Low Mad High Low Mad High

Cooking 826 585 503 854 797 576 833 826 489

Water heating 20 91 340 105 367 43 422

Space heating 653 324 157 91 98 57 70 131 89

TOTAL ENERGY 115 102 83 91 79 59 95 76 82 (GJcapita)

Source Guzman (1982)

As one would expect substantial national and local variations can be found For example in rural East Africa Openshaw [1978J has suggested a general pattern for the use of biomass fuels in which cooking accounts for 55 water heating 20 space heating 15 and ironing protection from animals and other minor uses 10 A recent national survey in Kenya [CBS 19801 supports this breakdown but also reveals large regional differences especially for space heating Shares for cooking and water heating range from 79-92 Space heating shares are as low as

- 59 shy

4 in Nairobi and the coastal region and as high as 20 in the cooler Rift Valley

In six low income villages of South India where space heating needs are negligible there was little variation in end-use shares the cooking share was 76-81 water heating 14-19 and lighting by kerosene and some electricity 2-3 [Reddy et ale 1980] In contrast in the much cooler climate of Chile a survey of eight subsistence villages found that the cooking share was 42-55 and space heating 23-52 [Diaz and del Valle 1984] Water heating absorbed 14-22 (except for one village with 6)

noting Several points related to estimates of this kind are worth

a Most survey information on end-uses is not given in terms of energy shares but of the proportions of households which use certain fuels to satisfy different end uses Data of this kind cannot be used to accurately estimate actual consumption for each fuel or end-use This is especially true where many households use multiple fuels for specific end-uses such as firewood and kerosene for cooking

b End-use consumption is often difficult to define because one end-use device frequently provides several end-use services As discussed in Chapter I the cooking fire often serves as the only source of space heating water heating and in many cases lighting

c The use of energy for income-earning activities is often great and may not be distinguished from pure household demand or may simply not be measured Examples include beer or spirit making boiling sugar from cane pottery tobacco and copra drying blacksmithing and baking Often these goods are produced for own-consumption and for sale The scale of errors that can arise if these energy uses are not measured or allocated correctly is well iHustrated by a rural survey in Bangladesh [Quader ampOmar 1982] For landless families annual consumption for all kinds of cooking and food preparation was 69 GJyear of which 66 GJ was for domestic cooking The small remainder was for parboiling rice and making ghur or sugar syrup For the largest farmers the equivalent figures were 163 and 83 GJyear The latter used more than twice as much fuel in total but little more than the landless poor for domestic cooking

d Religious festivals celebrations burials and other occasional functions may consume large amounts of fuel but be missed by energy consumption surveys

- 60 shy

F SUMMARY

Thi s chapter has reviewed many aspects of household energy consumption including data sources that might be utilized for national assessments ranges of energy consumption according to major variables energy use for specific tasks and methodologies for using these data in national assessments

The chapter purposefully avoided presenting typical consumption data that might be adopted in countries or locations where this information is needed but is lacking because household energy supplies and uses are almost invariably location-specific This is true of total consumption the mix of fuels employed and end-uses Within countries these differences are normally very large While the chapter has presented a number of examples of the range of data found in surveys there is no substitute for collecting or searching for household energy data that apply to the specific location in question

- 61 shy

CHAPTER III

ENERGY END-USES AND TECHNOLOGIES

This chapter examines household energy from the viewpoint of specific end-uses and the technologies which provide services such as cooking heat space heating lighting and refrigeration Its principal objective IS to present technical and economic data on end-use technologies such as the efficiencies costs and possible energy savings from using improved cooking stoves and lighting equipment

Section B examines energy for cooking and Section C discusses cooking stoves These are the largest sections of the chapter due to the importance of cooking energy in most developing country households

Sections 0 E and F examine lighting refrigeration and space heating respectively Although some of these services consume significant amounts of energy only in middle to high income households they are important to examine because they consume electricity are growing very rapidly in many developing countries and have a large potential for energy savings at relatively low cost

B COOKING

The amount of energy used for cooking depends on many factors the type of food cooked the number of meals cooked household size the specific combination of fuel and cooking equipment employed (type of stove cooking pans) and the way in which cooking devices are used

Consumption Ranges

Staples and other foods vary greatly in the amount of cooking time required and the rate of heat input For example rice is usually boiled or steamed for 20-30 minutes while kidney beans may be boiled for four hours or more Other foods are baked grilled or fried etc Table 31 presents some data from field measurements on the specific fuel consumption (SFC) to cook various staple foods The range of SFCs is about 7-225 MJkg even though woodfuel was used in all cases

- 62 shy

Table 31 Specific Fuel Consumption for Cooking Staple Foods (MJkg cooked food)

Rice Thai land 10 villages N India low incomes

high incomes ~I India Ungra village India 6 vi I I ages

Bangladesh Sakoa vi I I age Bangladesh 4 vi 1I ages Sri Lanka 1 vi 1 I age 21

(par-boiling rice)

Other To Upper Volta Beer Upper Volta Tortilla Mexico Kidney beans Mexico

Range of Mean Averages Source

158 122 - 229 Arnold ampde Lucia 11982) 214 16 - 27 NCAER 11959) 417 32 - 49 NCAER [1959] 248 Reddy (1980) 280 215 - 336 Reddy [19801

307 266 - 377 Quader ampOmar (19821 337 Quader ampOmar (1982] 38 Bialy 119791

(114) Bialy 119791

7 Sepp et al (19831 21 Cece I sk I 11984 ) 38 Evans 11984)

225 Evans [19841

al Range is for averages for six Sites including cooking other than for staple foods hence greater consumption at high incomes

bl Abundant firewood close to v i I I age bull

- 63 shy

Since diets include food other than staples another useful indicator is cooking energy consumption per person-meal or per personshyday Table 32 compares cooking fuel consumption per capita on a daily basis and is also based on field measurements Despite a wide range of locations and conditions the range of consumption is quite small In all cases food is cooked predominantly by open wood fire lower figures apply to efficient wood (or charcoal) stoves and modern fuels 1

Table 32 Specific Fuel Consumption for Cooking (MJcapitaday)

Household Percent Location Size MJcapday Biomass Source

F I j I 14 vi II ages 116 - 169 100 Siwatibau [1961 J I ndones I a Lombok 69 - 71 123 - 153 64 - 96 Weatherly [1960 J Bangladesh rural 137 95 Mahmud amp I s I am [19821

Indonesia Klaten 54 - 55 148 - 214 57 - 100 Weatherly [19801

S Africa Mondoro 15 I 100 Furness (1961] India Tamil Nadu 159 - 241 97 - 99 A I yasamy (1982 J Indonesia Luwu 56 - 63 170 - 244 99 - 100 Weather 1y (1960 I Bangladesh Sakoa 41 - 110 170 - 268 100 Quader ampOmar (19621

S Africa Chiwundra 175 100 Furness (1981) F i j I ato I Is 181 100 Anon 119821 Bangladesh Ulipur 186 100 Br I scoe (1979) India Karnataka 195 - 238 100 Reddy [1980)

India 2 villages 208 - 493 96 - 97 Bowonder amp Ravishankar (1964)

Bangladesh 4 villages 222 100 Br I scoe (19791 Mexico 2 villages 248 Evans (1984) India Pondlcherry 271 - 293 97 - 91 Gupta ampRao (1980)

]) In the industrialized countries where modern cooking fuels and equipment eating away from home and the use of partially cooked processed foods are almost universal specific fuel consumption for cooking in the late 1970s ranged from a low of 09 MJcapitaday in Canada to 29 MJcapitaday in the United Kingdom [Schipper 1982] These low figures may also be found in developing countries among single professional people

- 64 shy

The effect of different cooking technologies and variations in the type of meal cooked can be seen in Table 33 which is based on field tests in Fiji [Siwatibau 1981] Using as a point of reference the energy used for the second type of Indian meal using a kerosene primus stove some appliances have a consumption range of about 2 1 for different meals With other appliances there is little variation according to meal type The largest variations are for the type of appliance with a range of 141

Table 33 Fuel Consumption Relative Efficiencies and Cooking Times for Different Meals and Types of Cooking Appliances

Type of Cook Ing T~pe of Meal Appl iance Fijian Indian 1 Indian 2 Chinese 1 Chinese 2

EnerSl Consumption (MJ)

Kerosene primus 36 35 25 50 56 wick 121 61 82 52 69

Charcoal stove 133 140 131 151 199

Wood open fire 236 244 180 193 133 chulah 3~0 426 350 409 639 chanalan 210 250 195 199

Relative EnerSl Consumption ~rW~ l~In~_~) c~-Kerosene

primus 69 71 10 50 45 wick 21 41 30 48 36

Charcoal stove 19 18 19 17 25

Wood open fire bull11 10 14 13 19 chulah 07 06 07 06 04 chanalan 12 10 13 13

Cook in9 TI mes (minutes) Kerosene

primus 58 57 70 57 130 wick 59 55 63 60 147

Wood open fire 63 61 70 73 30 chulah 90 87 95 81 100 chanalan 75 67 88 81

Source Siwatibau (1981)

- 65 shy

Fuel Preferences

Cooking is an end-use in which one finds strong and often highly specific fuel preferences The reasons for choosing particular fuels and cooking appliances include ease of handling and lighting flame quality and temperature ability to secure fire from young children smokiness and the taste imparted to food as well as relative prices and availability of fuels These same factors may lead households to reject improvements such as more efficient stoves which do not satisfy their customs and preferences Some examples of these preferences and thei r weight in decisions regarding fuel choices are given below

In the town of Waterloo Sierra Leone al though the average family spent 30 of its income on firewood two thirds of them would not switch from it for any reason whatsoever The other third were prepared to change to charcoal or at worst kerosene The reasons for preferring woodfuels included food tastes safety and the wider range of cooking methods that are possible with an open fire The cost of woodfuels relative to that of fossil fuels was the least important consideration [Cline-Cole 1981]

Protection against shortages of modern fuels is another key factor often expressed by the ownership of more than one type of fuelcooking device In urban areas of the Philippines for example wood and charcoal are kept as emergency fuels in case gas and electricity supplies fail [PME 1982] Multiple fuel use is also common for different cooking tasks Many surveys have found that woodfuels are used primarily for cooking staples which may take on an oily taste on a kerosene stove while kerosene is strongly preferred for quick snacks or boiling small amounts of water for hot drinks as in Indonesia [Weatherly 1980]

In summary it is difficult to generalize about consumption levels or fuel and equipment choices for cooking Where interventions are being considered local quantitative and attitudinal information must be used as a basis

C COOKING STOVES AND EQUIPMENT

Since much already has been written on the problems and successes of improved cook stove (rCS) programs [Foley amp Moss 1983 Joseph amp Hassrick 1984 Manibog 1984] this section will not review these programs Nevertheless it is worthwhile to note the important questions which these programs indicate should be asked in considering any improved stove program (1) What improvements do consumers want (2) Does the improved stove provide them in the consumers jUdgement (3) Will the stove save fuel and (4) What does it cost

It is critical that stoves be designed and disseminated around social preferences as well as technical factors Stove users producers

- 66 shy

disseminators developers and evaluators should all be involved in any stove development and dissemination project since each group has its own set of objectives priorities and measures of success Successful stove design is largely a matter of striking the right compromise between these values particularly those of the users The active participation of women extension groups and stove producers has proved to be essential to the success of stove programs [Joseph ampHassrick 1984]

Before discussing stoves we must note that they are only one part of the cooking system Other factors such as the type of cooking pot how well pots fit the stove openings whether lids are used and management of the fire and fuel are important to fuel and cost savings and social acceptability Table 34 lists these factors and describes how they affect energy efficiencies and fuel savings

Table 34 Factors Affecting Cooking Efficiencies

Giving Higher Efficiencies Giving Lower Efficiencies

Fuel --dry wood dry c I I mate - wet wood moist climate

small wood pieces - large wood pieces (uneven and sometimes (even air to fuel ratio) inadequate air to fuel ratio) dung and

crop residues (usually higher moisture content)

Fuel Use and Cooking Site careful fire tending - poor fire tending (burning rate to match required (eg attention to other domestic power output for cooking task tasks) fire alight for minimum periods before and after cooking) indoor cook Ing - exposed outdoor site (but see text on (protection from drafts) smoke and health effects)

Stove and Equipment alUMinium pots - clay pots (good heat transfer) use of pot I Ids - no pot I ids (reduced heat losses) large pot small firestove - smal I pot large firestove pot embedded Into stove opening - non-embedded pot (large heat transfer area) well-fitted pot(s) with sma I I gap - poorly fitted pot(s) between pot and stove body (increased heat transfer) new stove good condition - old stove poor condition (eg reduced heat loss through cracks) metal ceramic-I ined stove - clay or mud stove open fire

Cook In9 Methods stove well adapted to or allows - stove ill-adapted to customary Improvements in methods methods food preparation to reduce cooking - no Initial preparation times (eg pre-soaking of cereals beans) use of ancill iary equipment (eg hay box for extended slow cooking thus reducing need for stove)

- 67 shy

Stove Types

A summary of stove types and their advantages and disadvantages is presented in Annex 5 [Prasad et a1 1983] This section presents only general comments and ranges of technical data

Improved Cook Stove programs initially focused on rural mud and clay stoves usually to be built by the intended user They generally had poor performance and acceptance (see Annex 5 for their main disadvantages) More recently attention has turned to urban and perishyurban consumers to ceramic and metal stoves for burning wood or charcoal and to construction by artisans with distribution through the market perhaps with government subsidies Acceptance has improved in some cases dramatically Quite rapid increases in stove production and sales are now being seen in several countries

For example in Kenya some 84000 improved Jiko stoves costing $4-6 have been sold in a period of 24 months [Hyman 1986] In Niger about 40000 scrap metal woodburning stoves costing less than $6 have been sold in 24 months [UNDPThe World Bank 1987] And in Nepal a concerted effort is being made to introduce improved woodstoves as part of a World Bank Conununity Forestry Development and Training Project Over 10000 stoves (mainly ceramic-insert and double-wall design) had been installed by 1985

Stove Efficiencies and Fuel Savings

Stoves are usually rated and compared to traditional cooking methods in terms of efficiency (see Chapter I for definitions) Other important user criteria are the maximum and minimum power output ie output range and turn-down ratio the type of fuel including the size and uniformity of firewood pieces equipment lifetime and cost

Early emphasis on achieving high efficiencies often ignored the other technical aspects which are equally important for designing acceptable and convenient stoves [Prasad et a1 1983 Manibog 1984] However some compromise between the various technical factors is inevitable in designing a new stove For example efficiencies are often extremely low at low power outputs but to correct for this (by altering the air flow to the combustion chamber) may upset the power range and efficiencies at higher power outputs

Information on basic construction designs and technical details such as efficiencies power ranges and labor and material needs for specific improved clay mud ceramic and metal stoves can be found in de Lepeliere et ale [1981] de Lepeliere [1982] Prasad [1982] Prasad amp Sangren [1983] Sulitlatu Krist-Spit amp Bussman [1983] Strasfogel [1983 ab] Baldwin amp Strasfoge1 [1983] Prasad amp Verhaart [1983] and Foley amp Moss [1983]

- 68 shy

As a result stoves with high efficiencies in laboratory tests have failed to produce the expected fuel savings under practical conditions This is usually because cooks prefer (or are forced) to operate the stove in ways that are sub-optimal for maximum efficiency in order to make up for various technical deficiencies Alternatively cooks may simply be wasteful in their use of fuel For example a stove may be filled to the brim with fuel which is allowed to burn out completely long after the cooking pot has been removed

On the other hand improved stoves which have been designed taking into consideration users habits have been shown to save substantial amounts of fuel under real life conditions For example in Senegal metal stoves consistently achieved fuel savings of about 30 compared to open fires when used for the same meals and cooking environment as predicted by laboratory tests [Ban 1985]

As this example suggests it is essential to compare like with like when assessing stove performance The failure to do this underlies much of the controversy and conflicting evidence on whether an improved stove is more efficient or needs less fuel than a traditional stove Much of this controversy can be ascribed to (l) comparing different products eg a one-pot and two-pot stove [Bialy 1983] (2) using different cooking utensils eg aluminium versus clay pots (3) using different test procedures and (4) poor definitions of test procedures Given these disparities it is no wonder that widely different efficiencies are reported in the literature even for the same type of stove [Gill 1983]

To clear up this confusion standard efficiency tests have been devised and are being used more and more [VITA 1984] See Annex 6 on Stove Performance Testing Procedures These tests do not measure efficiency in the narrow technical sense (ie utilized heat outputfuel energy input) but rather the Specific Fuel Consumption (SFC) for a defined cooking cycle such as preparing a standard meal (see Table 32)

The wide diversity in efficiency values is depicted in Table 35 which provides a set of cooking efficiencies that can be used as reasonably reliable broad guidelines Nevertheless actual measurements of fuel use per cooking cycle yield superior values and should be used in place of these guidelines whenever they are available The efficiencies provided in Table 35 are based on a variety of sources Before applying these values one should be aware of the factors which influence cooking efficiencies and SFCa shown in Table 34

- 69 shy

Table 35 Average Cooking Efficiencies for Various Stoves and Fuels a (Percent)

Acceptab I e ~ FuelStove Type Lab b Field

=c Value

Wood Open fire (clay pots) 5 - 10 7 Open fire (3 stone 18 - 24 13 - 15 15

alulllinum pot) Ground oven (eg Ethiopian altad 3 - 6 5 Mudclay 11 - 23 8 - 14 10 Brick 15 - 25 13 - 16 15 Portable Metal Stove 25 - 35 20 - 30 25

Charcoal ClaYlaud 20 - 36 15 - 25 15 Metal (lined) 18 - 30 20 - 35 25

Kerosene Wick

Multiple wick 28 - 32 25 - 45 3 Wick Single wick 20 - 40 20 - 35 30

Pres sur i zed ( 0U ) 23 - 65 25 - 55 40

Gas (LPG) Butane 38 - 65 40 - 60 45

Electricity Single element 55 - 80 55 - 75 65 Rice cooker 85 Electric jugpot 80 - 90+ 85

a Assuming aluminum cooking pots unless otherwise indicated b Mostly from water boiling tests c Generally reflects cooking cycle tests ~ Acceptab Ie assum i ng that the dom i nant stove types are higher qua I i ty

eXaRples of the type ie excluding stoves demonstrated as having inferior eff icienc les

Other Technical Aspects

Reliability and longevity are also important design aspects In measuring longevity the half-life concept is often used in the Ies literature [Wood 1981] This refers to the number of years after which half the stoves that were originally disseminated are no longer in use

- 70 -

Smokiness and its relationship to eye irritations eye disease chest complaints and other afflictions among women (or other family members) has often been neglected by stove designers and analysts Nevertheless it is an important criterion in stove acceptance Recent work by Smith et al [1984] in different areas of India suggests that smoke from cooking fires can be highly carcinogenic and that carcinogen levels greatly exceed acceptable exposure rates in developed countries Evidence of correspondingly high carcinoma incidence in housewives is still slim however On the other hand smokiness is sometimes seen as a benefit since it repels insects and the smoke has creosotes which preserve thatch and timber roofs from premature deterioration

Stove Costs

Although serious work on stove programs has been going on for five years there still is very little economic data available for different types of stoves It is not always clear in this data whether costs apply to the stove only the fuel only or the stove and fuel Initial costs andor lifetimes also may not be given so that payback periods cannot be calculated Furthermore costs to the stove user may be estimated but costs for other essential groups in the design production and dissemination chain are frequently neglected To the producer (artisan or stove owner) the important economic factors are profits or the return to labor to the stove developer the development and testing costs and to the disseminating agency the margins after accounting for the costs of marketing distribution training monitoring and possibly subsidizing the improved stove All these costs and margins should be considered since an improved stove program can fail if the economics are poor for anyone link in the chain

The costs of stoves vary widely by type technical specification (size quality of materials and workmanship etc) and country The costs of woodburning stoves can range from less than $100 for a simple scrap metal type in some developing countries to as much as $60 for a modern heavy metal oven Experience in a number of countries indicates that improved wood and charcoal burning stoves can be produced and sold for anywhere from US$1 to US$15 For example in Kenya the very successful improved Jiko -- a charcoal stove of metal ceramic construction -- presently sells for U8$4-8 while in Ghana local scrap metal woodburning stoves cost about U8$1 and heavy metal stoves sell for about U8$5-8 In Peru an improved ceramic stove costs about U8$1-2

While prices may vary considerably from country to country within a country there tends to be a relationship between the prices of the different types of stoves This relationship is summarized in Table 36

- 71 shy

Table 36 General ized Stove Cost Index (mud stoves =base)

Woodburnlng Stoves

Mud 10

Clay 15 20

Metal 060 - 600

Charcoal 10 25

Kerosene 2 - 8

Gas 120

Electric 140

To the user the amortized cost of an improved stove would normally be a minor factor in the total lifetime of the stove But the investment to purchase the stove occuring at one point in time may be a major deterrent to poor families For the user the economics of an improved stove is determined by the amount of fuel saved and if adoption demands a switch in fuel relative fuel costs

This point is clearly illustrated by the recent cost comparisons of eleven stovefuel combinations in Thailand presented in Table 31 The amortized cost of the stove ranges from about 13 to as little as 05 of the total monthly costs including fuel The total monthly costs are dominated by the unit costs of the fuel and by the efficiencies

For this reason the most useful cost indicator for stove users is the payback period ie the time required to pay back the investment on the stove (plus any repair costs) through reduced fuel costs Methods for estimating payback times are presented in Annex 7

Payback periods as short as 13 days have been reported for an improved charcoal stove plus a change to aluminium pots at current market prices in Ethiopia [UNDPWorld Bank 1984b] Payback periods of one and three months have been estimated respectively for metal stoves in Burkina Faso [Sepp et al 1983] and ceramic stoves in Nepal [Bhattarai et al 1984] In contrast heavy mud stoves built in situ by artisans have had payback periods of as long as 12-30 months

- 72 shy

Table 37 Efficiencies and Total Costs of Various FuelStove Combinations in Thall and

Stove Fuel Cost Stove Cost Total Cost Fuel Type Eff Ic lency per Kg per Month per Month per Month

Rubber Wood

Rice husk

Sawdust

Charcoal

Corn cob

Rice husk log

Sawdust log

Bucket

Rangsit

2-hole mud

l-ga I can

Bucket

----------------------baht-------------------shy

24 16 114 16 130

23 16 119 16 135

16 19 204 30 234

12 19 261 22 266

16 76 576 03 564

18 1 70 646 16 662

14 170 884 16 900

21 145 893 16 909

17 145 1124 16 1140

25 185 1267 16 1283

18 203 1892 16 1908

Source I s I am et a I [1984)

Dissemination and Impact

In addition to stove costs and payback periods any stove program must also allow for regional fuel constraints user preferences and institutional requirements Manibog [1984] discusses thoroughly the problems of carrying out Ies projects There are six essential conditions for getting operational stoves into widespread use These include (1) active participation of women (stove users) artisans and the marketing or disseminating (eg extension) workers in developing or adapting a stove design (2) proof that long-run market production delivery and maintenance systems exist or can be established (3) establishment of training programs for local artisans or extension workers (4) development of and strong financial support for a strategy to market the chosen stoves and appliances based on comprehensive

- 13 shy

acceptance surveys and possibly incentive pricing systems to stimulate early adoption of the new technology (5) continued support for research and monitoring of stove development and (6) market conditions which allow competitive models to be developed and reach the market

The potential gains from improved woodstove programs are enormous Many of them do not relate directly to energy but involve for example better health and hygiene safety for young childern and improvements to the general cooking environment At the same time reductions of 30-50 in fuel use can be achieved and should be easier to deliver and manage and in less time than supply-side developments such as fuel plantations

The cumulative impact of an improved stoves program on national fuel savings can be significant As explained in Tropical Forests A Call for Action [WRI 1985] this impact will depend on the number of households that use the stove the amount of time the stove is used and the actual gains in efficiency obtained from the stove For example if 50 of households in a region use improved stoves for cooking 80 of their meals and the stoves double the cooking efficiency a 20 decrease in fuelwood consumption would be achieved However if only 10 of the households in a region use the stove and cook only 50 of their meals on it the decrease in fuelwood use for cooking is only 25 for the region

A recent study in the Kathmandu Valley Nepal -- a region containing some 800000 people -- estimated that improved stoves could save up to 92000 tons a year of fuelwood valued at US$6 million This is equivalent to the annual yield from a 14000-hectare fuel wood plantation in local conditions

D LIGHTING

Although lighting uses relatively little energy it has an important place in household energy for three reasons First lighting usually involves the use of commercial energy and often is the only use for such energy by poor households Second low and middle income families view improved lighting as a high priority in the achievement of better living standards Third for poor families improved lighting usually involves substantial equipment costs whether they be for a kerosene pressure lamp or electric light fittings and connection charges

As a result energy consumption for lighting normally increases quite rapidly with income above a certain threshold level but at the same time may be a critical component in the energy budgets of the poor Consumption is also highly dependent on energy prices and technologies which have a very large range of end-use efficiencies and hence a large potential for energy savings without sacrificing lighting standards

- 74 shy

Although information on energy use for lighting has improved with recent surveys in general it has been poor Household surveys often fail to separate consumption of electricity and liquid fuels (eg kerosene) into lighting and other end-uses and very few studies have followed the energy used for lighting through to the ultimate level of service provided such as levels of illumination and daily hours of lighting

Measurement Units and Standards

The basic unit of light intensity is the lumen Um) which combines a physical measure of the light level with the response to this by the human eye Another unit is the lumenWatt UmW) which introduces measures both of efficiency and the rate of light output over time For instance a 100-W incandescent bulb typically provides 15-18 lmW or a luminous flux of 1800 lumen Illuminance refers to the effective light level per unit area and is the measure on which lighting standards are set An illuminance of 1 lumenft is equal to one footcandle Table 38 provides international lighting standards which were devised for developed countries They suggest that some working conditions require a lighting intensity seven times greater than normal background lighting However these standards are often too high to be considered practical for developing country applications where incomes are low andor electricity costs are high eg for home or village street lighting

Table 38 Lighting Standards for Various liousehold Activities

Activity IES Standard (footcandles lumenft2)

Passageways relaxation and recreation 10

Reading (book magazines and newspapers) 30

Working (kitchen sink handwriting study) 70

~ Leckie J bullbull ed 119751

Lighting Energy Fuels and Technologies

Many poor families in developing countries rely on the cooking fire and possibly candles and sparing use of an electric torch to meet all their lighting needs For others electricity and kerosene are the

- 75 shy

main energy sources for lighting Of these electricity is usually preferred (although it may not be available or is too expensive) because of its cleanl iness convenience and better spectral light quality Kerosene or benzine lamps on the other hand have a high glare factor are hot and in the case of pressure lamps are very noisy Many electrified households however consume significant amounts of kerosene as a supplementary lighting source andor during power shutdowns Benzine is often used instead of kerosene by higher income households in non-electrified villages Gas lighting is a rarity

Table 39 indicates the range of kerosene consumption for lighting based on the few surveys where this end-use was distinguished and where 90-100 of lighting needs were met bJ_~~rosen For Jow to middle income groups consumption is roughly 6~i~ers 18 ~~ ~jb per household per year or about 007 - 028 liters per nig t -althougn much

~(s--MJ(lt~ f 14l) Table 39 Household Kerosene Consumption for Lighting

(liters per year)

Kerosene Mean Range Source

Rural

Bangladesh Sakoa low income high income

India Balagere Bhogapuram 6 villages

all rurallow income all ruralhigh income

Indonesia 3 villages SUMatra all rural 1976

Pakistan all rurallow Income

Sri Lanka

Thai land

India a II urbanlow Income all urbanhigh income

Indonesia 1976

28 143

35 42 52 45-61 25 51

70-500 254 148

34

104 96-140

55-91

31 86

570

Quader ampOmar (1982

Bowonder amp Ravishankar (19841 Reddy [1980 1 NataraJan 11985]

Weatherly 119801 Down 119831 Strout 119781

FBS [1983

WiJeslnghe (1984)

Arnold deLucla ( 1982)

Natarajan (19851

Strout (19781

- 76 shy

higher figures have been reported for Indonesia possibly because of exceptionally low kerosene prices at the time Lighting periods in these surveyed households were typically about 2-4 hours per night

Table 310 presents data for India on the consumption of lighting kerosene and electricity by income level urban-rural differences and whether houses are electrified or not [Natarajan 1985] Notable points are that consumption increases significantly with income only above annual incomes of around 6000 rupees (approx US$600) and kerosene 1S used rather extensively in electrified households especially in rural areas The substitution ratios shown in the final column are discussed below

Kerosene and benzine are burned either in open wick lamps (typically with a naked flame from a wick protruding from a simple jar or bottle of fuel) enclosed wick lamps in which the wick is surrounded by a glass chimney that creates an updraft past the wick and promotes a

Table 310 Energy Use for lighting in Electrified and Non-Electrified Households India 1979

(by Income and Urban-Rural location)

Annual Income Non-Electrified Electrified Substitution (thousand Kerosene Kerosene Elee Total Ratio ~ Rupees) (iltres) GJ (litres) (kWh) GJ ( I i treskWh)

~ lt3 3- 6 6-12

12-18 18 All

Urban lt3 3-6 6-12

12-18 18 All

25 29 41 46 51 28

29 31 31 50 86 31

087 102 144 160 179 097

103 107 107 174 302 108

90 84

104 101 106 91

45 61 48 39 39 53

156 163 205 283 322 178

164 189 243 324 425 217

088 010 088 013 110 015 137 013 153 013 096 011

075 015 089 013 104 011 130 014 167 019 096 012

Substitution ratio is the difference In kerosene use between non-e I ectr if jed and electrified households divided by electr Icity use in the latter (Iitres kerosenekWh electricity per year)

~ NataraJan [19851

- 77 shy

hotter brighter flame or pressurized lamps which normally employ a coated mantle to provide an intense white light

Table 311 provides data on light intensities and the specific fuel consumption of kerosene lamps Comparing this with Table 37 it can be seen that most kerosene lamps provide very low lighting intensities far below those required to meet the illumination standards accepted in developed countries Indeed in a survey of low income Indonesian households Weatherly [1980] found that the simplest small wick bottle lamps although burning only 10 millilitres of fuel hourly gave out a light equivalent to only a 2-Watt electric torch bulb

Table 311 Technical Characteristics of Lighting FuelLamp Combinations

Fuel and Light Intensity Fuel Use Consumption Lamp Type (Foot candles at 30 em) (millilitrehour) Index a-Kerosene

Mean Fishcan and wick 05 98 127 Standing 15 up to 4 120 52 Hurricane 3 1 - 35 121 26 Pressure (Ti I I y) 32 20 - 70 478 10

Benzine Pressure (Coleman)

badly pumped 20 8 - 25 486 15 well pumped 25 20 - 45

Electricity 60-W incandescent 40 (60 Wh)

a Consumption index is measured as power input per unit I ighting intensity normal ized to 1 for the 60-W bulb Calorific values used are kerosene 35 MJliter benzine 33 MJliter electricity 36 MJkWh

Source Siwatibau 19811

The costs of various lighting technologies are given in Table 313 For the poorest families these costs are a major deterrent to adopting lighting standards which improve on simple wick lamps However for families who own or are choosing between relatively advanced lighting equipment initial costs are a small part of total life-cycle costs

Relative efficiencies and energy prices are therefore critical components in the economics of lighting Here it is worth noting that in the Indonesian case just cited the respective power inputs were 001 literhour x 35 MJliter = 35 MJhour for the kerosene lamps and 0002 kW x 36 MJkWh = 0012 MJhour for the 2-W electric bulb with the same lighting intensity Thus the wick lamps were roughly 50 times less efficient than incandescent electric lighting Few kerosene lamps have

- 78 shy

an efficiency better than 1l0th that of electric lighting as can be seen in the final column of Table 311 which gives an index of power input per unit lighting intensity As a result one frequently finds that the running costs of electric lighting are less--or much less--than lighting by kerosene for an equivalent light output

Table 312 lamp Costs

Country Type of lamp Cost 1984

(USS)

Fiji large Kerosene large Benzine Small Benz i ne

45 43 29

liberia Small kerosene (Chinese) Medium It It

large It

550 750

1175

This point is of great importance for fuel substitution Since electricity almost invariably replaces kerosene for lighting and not vice versa one might expect energy consumption to fall after the switch due to the much greater efficiency of electric lighting However most consumers increase their lighting standards (intensities) at the same time

The important quantity for analysts therefore is the actual energy substitution ratio This can be established only by comparative surveys of electricity and kerosene users at similar socio-economic levels or preferably by consumption surveys before and after the substitution is made The results from the few analyses of this kind that have been made are given below

In Klaten Indonesia Weatherly [1980] found that one kWh of electricity for lighting replaced 051 liters of kerosene an electricitykerosene energy ratio of 3618 MJ or 15 In six South Indian villages [Reddy 1980] electrified households used one kWh for every 015-028 litres of kerosene in non-electrified households an energy ratio of 115 to 127 In the Indian survey reported in Table 39 the ratio for the bulk of rural and urban households was a bit lower at 013 - 015 litres per kWh an energy ratio of 113 to 115

Table 313 presents the costs and specific consumption of electric luminaires which include incandescent bulbs standard fluorescent lamps and advanced technologies available in the early 1980s The costs are for retail markets in Brazil in 1983 converted to US dollars One notable point is the large range in lighting

- 19 shy

efficiencies expressed here in lumen output per Watt input The range is from 12 to 63 lumenwatt a ratio of 51 The second point is the much higher cost of the fluorescent and advanced devices although these are offset by their much longer lifetimes

For consumers the economics of these lighting methods depend onmiddot the tradeoff between the high costs of efficient equipment and the lower running costs of this equipment The economics can best be compared by estimating payback times as with stoves (see Annex 1) A payback calculation to compare the 40 W incandescent bulb to the 16 W fluorescent light normalized to an output of 1000 lumen is presented in Table 314 Despite the 18-fold difference in equipment cost the total costs over the first 5000 hours when the fluorescent light has to be replaced are very similar at around $11 for an electricity price of 3 USckWh For any higher electricity charge the fluorescent light would be the most economic on a life-cycle basis

Table 313 Technical Characteristics and Costs of Electric lighting Technologies

(Market Prices in Brazil 1983)

light Specific Equipment Technology OutpuT Consumption li fe Cost ampPower Input (lumens) ( I umenwatt) (hours) (USS 1983)

Incandescent

40 W bulb 480 60 Wbulb 850

100 W bulb 1500

Fluorescent tubes

11 Wtube 400 16 Wtube 900

Advanced fluorescent bulbs

9 W bulb 425 13 W bulb 500 18 W bulb 1100

High intensity discharge

55 W bulb 2250

al Including ballast costing US$4 with

~ Goldemberg et al (1984)

120 143 149

1000 1000 1000

357 556

5000 5000

476 385

625

5000 6000 7500

41 ~7 5000

life of 20000 hours

05 05 06

130 al 130 al

130 92

250

120

- 80 shy

Table 314 Payback Analysis for 16 W Fluorescent Lighting Compared to 40 W Incandescent Bulbs

(data from Table 312)

For light output of 1000 lumen and lighting for 5000 hours 40 W bulbs 16 Wfluorescent

Lumen per unit No of units required Lifetinae per unit (hours) Unit cost (USS)

Equipnaent costs for 5000 hours Units purchased Equipment costs (USS)

Energy costs general Watts per 1000 lumen output kWh for 5000 hours lighting

Total costs at 3fkWh Equipment Electricity

TOTAL

Payback period approx infinite

Total costs at 5fkWh Equipnaent Electricity

TOTAL

480 900 21 11

1000 5000 05 130 a

102 11 51 143

83 18 415 90

51 143 ~ 27

17 55 17 0

51 143 2075 45

2585 188

Payback period approx 5000 hours x 1882585 = 3636 hours

727 days (2 years) if 5 hours lighting per night

a Includes bal last at USS4 Replacement required only after 20000 hours

Photovo1taic Lighting

Photovo1taic lighting in some instances can be a viable alternative to the more traditional lighting systems and therefore should be examined also A typical household solar lighting system consists of a solar panel or arra with an output capacity of 20-30 Watts for a solar input of 1 kWm (ie 20-30 peak Watts or Wp) a deep-charge battery and 2-3 fluorescent lights which are run for about four hours per night Outputs for TV and radio are often provided as well Total kit costs (i e panel lights battery and wiring) average U8$250-350 while total installed costs are about U8$300-400 (or about $12-15 per Wp) Panel costs were approximately U8$6-9 per peak Watt in 1984 for

- 81 shy

small-scale household systems but are expected to fall steadily These costs reflect favorable situations where good market transportation and installation conditions exist ie mostly in urban areas where grid electricity usually is available Although running costs are close to zero actual financial life-time costs cannot be generalized since they depend on the average level of solar radiation its seasonal as well as day-to-day variability and the amount of lighting demanded from the system However some estimates can be made as in the example below

Example

Assume interest (discount) rate = 10 10-year kit life ie amortization factor = 0162 total daily insolation equivalent to 1 kW for 5 hours

Then 30 Wkit costing $300 installed will produce 30 x 5 x 365 = 54750 kWhyear

Annualized cost of installed kit will be 0163 x $300 = $50

And thus elecric power cost produced with such a kit would be $5054750 = $09lkWh

Studies which have compared the economics of kerosene dieselshyelectric and solar lighting in remote rural areas tend to find that solar and diesel costs are fairly close and generally lower than kerosene assuming the same quantity of lighting for each method [Wade 1983] Although this is likely to be the case in sunny regions where no electric grid exists and diesel fuel is expensive or hard to obtain where these limitations do not exist photovoltaic lighting is unlikely to be economic -- at least at present costs In the absence of subsidies the high initial cost 18 bound to be an insurmountable barrier for most households

One should also recognize that the economics of all decentralized energy sources compared to those of centralized systems (eg grid distribution of electricity) depend on energy consumption levels Once the capital costs of grid extension have been met any increases in consumption are related only to generation costs while the costs of the distribution system per unit of consumption actually fall In contrast with a decentralized system each increment of energy use (or power) requires a complete additional supply unit For this reason it can often be shown that decentralized (eg solar) energy is competitive with grid power at low consumption levels but compares poorly at higher levels

- 82 shy

E REFRIGERATION AND OTHER ELECTRICAL END-USES

Higher income households normally consume substantial amounts of electricity for uses other than lighting The major demands are for refrigeration and air conditioning with minor amounts for TV radio and hi-fi ironing and electric power tools etc

The key parameters in assessing consumption are (1) ownership levels (and acquisition rates) of the major items of equipment (2) period of use (Le hours per day) and (3) specific consumption (ie kW per appliance) Since these factors can be estimated only by detailed measurements over long periods of time more practical indicators are given by typical ranges of consumption according to equipment ownership

Two examples of the way in which consumption increases as equipment is purchased are shown in Table 315 for Fiji and Sri Lanka In both cases the large increments in consumption occur when refrigerators and air conditioning are acquired

Table 315 Electricity Consumption by Appliance Ownership Fiji and Sri Lanka

Equipment Electricity Use Location Owned (kWhmonth)

F I j i Lighting o - 15 + iron amp radio 15 - 35 + refrigerator 35 - 150 + hot water ampwashing machine 150 - 300 + cooker amp air conditioning abOve 300

Sri Lanka LI ght i ng fan Iron 27 + hot plate ampkettle 190 + hot water ampwashing machine 280 + air conditioning 700

Sources Siwatibau (19811 Munasinghe [19831

To assess the economics and potential energy savings of conservation programs and other kinds of technology substitution the technical characteristics and patterns of using the existing equipment stock and possible replacements must be determined Very little information of this kind has been recorded for developing countries However the potential for improving energy efficiencies is undoubtedly large For example the specific consumption (Le Watts per liter

- 83 shy

capacity under standardized operating conditions of Japanese model refrigerators fell by a factor of 37 between 1971-73 and 1980 from 0618 Wlitre to 0166 Wlitre [lEE 1980J With air conditioning one also finds a range of about 3 1 between the most and least efficient technologies in current use

A number of attempts have been made to induce consumers to adopt some of the more energy efficient equipment that has been tried in developing countries These include labeling appliances for energy use and setting efficiency standards on domestic producers and imported equipment as well as controlling electricity pricing and tariff structures

F SPACE HEATING

The importance of space heating in some areas of developing countries has already been stressed Several surveys for example in Lesotho [Best 1979 and Tanzania [Skutsch 1984 have shown that it may as much as double the amount of energy used in winter as compared to summer The main impact of space heating is not only that it raises total fuel needs but also that it raises them during seasons when it is more difficult to collect store and dry biofuels

Despite this there is little information from which to determine where and when heating is a significant end-use what levels of consumption to expect or what might be done to reduce these needs Two reasons for this dearth of information stand out First as discussed before space heating is provided by any heat source in a dwelling and cannot easily be distinguished from other end-uses So there is little reliable information on specific consumption levels Second ambient temperatures are rarely reported in household surveys This means that there is little information on which to correlate space heating needs with easily measured or available quantities such as local weather data

A simple method for assessing space heating needs which is adequate for most analyses is provided in Figure 31 The promotion and economic analysis of methods to reduce space heating loads are much more difficult in developing countries than in industrialized countries This is primarily because the majority of dwellings are poorly constructed so that heat is lost by the infiltration of cold air through innumerable gaps in the structure and around doors and windows etc These are not so easily prevented as in well-constructed houses by weather stripping remedies Reducing conduction losses through the fabric of the dwelling by applying thermal insulation has considerable potential for saving energy in many areas but the idea is novel and there is usually no tradition of using these techniques

I shy

- 84 shy

FIGURE 31 Method of Estimating Space Heating Consumption from Total Energy Use and Ambient Temperature

Average Delivered

Energy for

Time c Period

B

High Low Temperature Temperature

World Bonk-31214

The graph plots total delivered energy consumption averaged over periods such as a day or week occurring within the living space The portion from A to B is for non-space heating end-uses At Point B heat is generated from these uses at the same rate that it escapes from the dwelling to the cooler external surroundings To the right of B as the external temperature falls the temperature inside the dwelling would drop unless extra heat is generated To maintain the internal temperature the occupants must therefore burn fuel at a higher rate The line B-C records this effect and allows for adjustments of internal temperature during colder weather For example if the occupants maintain a (roughly) constant average internal temperature--eg using a thermostat and central heating system the slope of B-C would be steeper than if temperatures were allowed to fall as the weather gets colder A few measurements of daily or weekly fuel use at different external temperatures can establish the position and slopes of the lines A-B and B-C Annual fuel consumption can then be estimated using temperature data for the whole year assuming that the dwelling is occupied More sophisticated methods can be found in many texts on heating and energy conservation in buildings

- 85 shy

CHAPTER IV

HOUSEHOLD ENERGY SUPPLIES

This chapter discusses household energy resources and supplies focusing on firewood charcoal and other traditional fuels used by households in developing countries The chapter does not discuss supplies of petroleum gas or electricity since there is much literature already available on these topics

As with consumption household fuel supply issues can be subtle and complex Where woodfuels are scarce and forests depleted the obvious answer would appear to be to plant more trees for fuel It However the many failures to do just this over the past decade underline the fact that there are rarely simple answers to the problems of woodfuel scarcity and indeed that people frequently have been misled by trying to answer the wrong questions

Experience to date suggests that fundamental questions must be asked before any effort to increase biofuel supplies is undertaken For example Is fuel scarcity really the problem For whom Is tree growing the solution Who wants to and can grow trees Are the main issues technical and economic or do they relate to management and social structures

Section B reviews some of the issues involved in household fuel use decisions and presents observations of behavioral patterns and characteristics of fuel users under various circumstances

Section C discusses fuelwood supplies providing data on yields characteristics of species and methods of analyzing production in physical and economic terms

Section D looks at transport and other marketing costs which strongly affect the incentives for producing fuelwood and the retail prices of wood in urban areas If producer prices are low farmers are unlikely to grow fuelwood and continued deforestation by low-cost cutting of natural woodlands may be inevitable Transport and other marketing costs also play an important role in the relative economics of wood charcoal and densified crop residues for urban commercial fuels These costs are also significant in determining the command area of urban woodfuel supplies

Sections E F and G discuss the key issues in supplying charcoal crop residues and animal wastes respectively For charcoal these issues include access to and rights over the primary wood resources and the costs and efficiencies of converting them to charcoal For crop

- 86 shy

residues the issues involve the amount of residues that can be safely removed from the soil the costs of collection and competition with nonshyfuel uses The section on animal wastes includes a brief discussion on biogas

B BACKGROUND PERSPECTIVES

The African Sahel has experienced widespread deforestation and fuelwood depletion over the past decade and has become a priority target for attempts by governments and aid agencies to plant trees for fuel Yet by 1982 despite expenditures of about US$160 million only 25000 hectares of fuelwood plantations had been established and most of them were growing poorly [Weber 1982]

Similar disappointments have been experienced in other regions Although there have been a few successes it is still not clear why those who appear to face acute fuel scarcity are so often reluctant to take steps to increase their traditional fuel supplies Questions such as this which relate to the socio-economic background of traditional fuel supplies are fundamental to understanding the remainder of this chapter They are addressed here briefly before the technical and economic aspects of traditional fuel supplies are discussed There the focus is on production at the farm and village level rather than on large-scale managed plantations since the former is most frequently misunderstood

Village Biomass Systems

Rural inhabitants produce and depend on biomass materials of all kinds food fibre grass and crop residues for animal fodder timber for sale or construction materials crop residues for thatching and making artifacts such as baskets and biofuels Most of these resources and the land devoted to their production have alternative uses (or an opportunity cost for anyone use) while the materials are frequently exchanged within the village biomass economy in complex and subtle ways

At the same time it is reasonable to generalize that where household fuels are in such short supply that they amount to a problem requiring intervention or significant adaptations there will be shortages of one or more types of biomass material This is so because scarcities of traditional fuels are generally most severe in areas of high population density (with strong pressures to produce more from each unit of land) and in arid or semi-arid regions where the productivity of all kinds of biomass is low These biomass shortages may be general or they may be confined to critical sub-groups such as the landless poor and the small farmer

- 87 shy

Whether general or localized biomass shortages usually call for an integrated approach to restoring supplies Particularly where agricultural residues and animal wastes are used as fuels and are in scarce supply (at least for some classes andor in some seasons) supplyshydemand balances and remedial actions cannot look only at the fuel aspect of biomass products If they do they are likely to produce sub-optimal answers or lead to projects which are rejected fail once implemented or actually damage some parts of the community For example if animal fodder is scarce planting trees for woodfuels on grazing land--or planting with species such as Eucalyptus which have inedible leaves-shycould deny essential fodder resources to some people Conversely a fodder and dairy development scheme might not only improve nutritional standards and incomes but also solve the fuel problem by freeing up biomass resources which can be burned without harm to other production or consumption activities This latter approach has been shown to be an effective remedy for traditional fuel shortages in semi-arid areas of India for example (Bowonder et a1 1986] It is unlikely that this would have been recognized in the more narrow scope of analysis commonly taken in an energy assessment

Access to Resources

Differential access to resources is another reason why integrated approaches are usually essential In most village societies there are not only large differences among sub-groups in obvious biomassshyrelated assets such as land and cattle ownership (both of which may provide fuels) but also subtler rights and dependencies concerning fuel collection These may include rights to graze on or collect fuel from common lands customs about scavenging crop residues after the harvest or crop processing (eg rice straws and husks) and traditions over partshypayment for labor in fuel materials instead of cash Generally as fuel shortages develop these traditions dependencies and rights are altered to the disadvantage of the weakest sections of the community

Similar arguments apply to one of the most common approaches to biofuel shortages the promotion of small-scale tree growing for fuel and other purposes eg social and community forestry Those with the most serious fuel problems are generally the people who are least able to grow trees landless laborers small farmers who lack labor and other inputs required for tree care and pastoralists who lack the traditions of crop and tree planting In many places land tenure constraints are fundamental barriers to growing trees Farm tenancy often with precarious rights to the land periodic reallocations of land ownership (as in Burkina Faso) and creeping land enclosure effectively destroy incentives that do exist for farmers to invest in the long-term enterprise of tree growing (or in soil and water conservation efforts) (Foley amp Barnard 1984]

In most of these situations changes in community attitudes to land holding and access rights are required before the majority of people can either grow trees themselves or benefit from tree growing by

- 88 shy

others Quite fundamental changes also usually are required in village power and control structures or in leadership and the trust that people put on the village elite Planting communal trees along roadsides canal embankments and on waste ground as well as in village woodlots has taken root in many places and with considerable success But this success requires a consensus in the community about the need to grow trees how to distribute the work of tree care and how to divide the benefits

Involving the People

The need for integrated appoaches to inherently complex and socially stratified systems leads to a critical question How are the systems to be understood The discussion above suggests that before any actions can safely be taken food fuel fodder and fertilizer balances need to be constructed furthermore that these balances must differentiate between groups such as large medium and poor farmers landless laborers the landless non-farm population and so on Some analysts believe that identifying the critical constraints or scarcest resources requires the use of approaches such as farming systems analysis which look at the linkages and conflicts around all the key resources land labor water food and feedstuffs fuel and fiber Remedies which may not be primarily directed to energy are then based on findings about the operation of the system

However this ideal approach if conducted mainly by outside experts is extremely time-consuming requiring much more than a rapid sectoral survey Furthermore outsiders almost inevitably try to separate and compartmentalize what they think are the relevant factors in order to find and impose pattern and structure in the search for solutions These dichotomies may bear no relation to the holistic view of the people on the ground--the insiders--who may well see different overlaps interrelationships constraints and opportunities

The close involvement of local residents therefore is not only necessary to avoid sub-optimal--or rejected or damaging--solutions it may also be the best way of finding shortcuts to successful remedies Local residents better than any outside visitors know how their system operates where it fails and needs improvement and usually what needs to be done if extra resources are made available to work with Local grassroots voluntary organizations frequently share this knowledge are trusted by the village community and have the social commitment and motivation to effect change as well as the knowledge and ability to invent new approaches In short close liaison with local residents and voluntary organizations is a much better guarantee of success than any amount of data collected for desk analysis

Tree Loss and Tree Growing

The massive loss of forest and woodland that is occurring across the developing world [WRI 1985J requires broad integrative

- 89 shy

thinking if its true causes are to be recognized and effective remedies developed In most places the main causes of tree and forest depletion are clearances for arable and grazing lands due to population growth migration and resettlement schemes slash and burn farming with overshyrapid rotation cycles due to population pressures overgrazing of young trees and supportive grasslands uncontrolled bush fires and commercial logging for timber in some areas

Demand for fuel may play a major part in deforestation in two broad cases The first is when tree loss has gone a long way and the local rural population must cut fuel from the few remaining trees Fue1wood cutting thus may play a part in the final stages of tree depletion [Barnard 1985 Newcombe 1984b] The second case is where the demands of urban markets for woodfue1s (firewood or charcoal) are sufficiently large andor concentrated in particular areas

In some cases tree clearance for agriculture can produce a temporary glut of woodfue1s thus lowering prices and encouraging greater consumption and the substitution of woodfue1s for fossil fuels When the glut comes to an end there may be a sudden onset of woodfue1 shortages and a rapid rise in prices Woodfue1 gluts have occurred recently in Sri Lanka due to the large scale forest clearances of the Mahawe1i Development Project and in Nicaragua where vast numbers of diseased coffee bushes have been replaced and land reform measures have allocated forest land to peasant farmers

Tree planting or more productive management of existing forest resources is obviously necessary if these trends are to be decelerated or reversed But it may not be sufficient if other causes of deforestation that have nothing to do with fuel demand are not also tackled If woodfue1 consumption were to drop to zero overnight deforestation in many countries would still continue on a significant scale because of factors such as land clearing and overgrazing [Barnard 1985]

In particular urban pressures on woodfuels can rarely be halted merely by growing trees The entire structure of woodfue1 markets fees and permits to cut wood and access rights to forests must almost invariably be adjusted as well A full discussion of the issues involved is beyond the scope of this section but a concise description of the impact of urban fuel demands is included in Annex 8 (Barnard 1985]

One also needs to consider the incentives for growing trees especially where the aim is to provide woodfuels Planting weeding watering protecting and caring for trees takes time and effort and conflicts with other priorities This is particularly the case in arid areas where fue1wood scarcity generally is most acute because the planting season for both crops and trees is short Farmers may be able to plant a few trees each year but if tree growing in any larger volumes interferes directly with food production or off farm wage earning activities it is unlikely to be undertaken [Hoskins 1982]

- 90 shy

Where private farmers do plant trees in large volumes fuelwood supply beyond their immediate needs usually has a low priority--even in regions of considerable fuel scarcity This is so because often no well established market and transport systems exist for fuelwood to make private farmers able to profit financially from fuelwood production In most areas of the developing world trees are grown for some combination of timber pulpwood building poles fencing material animal fodder fruit or nuts shade live fencing and hedging windbreaks or aesthetic reasons Firewood is seen as a useful by-product rather than a major justification for planting There have been numerous attempts to promote quick-growing firewood species which have failed almost completely and may well have hampered the growing of other species which would have produced firewood as a by-product [Barnard 1985 French 1981 Weber 1982]

Table 41 provides a checklist of the potential benefits from rural tree growing The range of benefits which includes both private as well as social benefits suggests that programs based on narrowly defined objectives such as wood fuel supply may greatly understate the real value of trees to rural dwellers

It is this discrepancy between private benefits and social benefits which creates the divergence between private and social incentives for tree growing From the farmers perspective the social costs externalitiesgt of not growing trees while continuing to deplete the already thinning forestry reserves or burning biomass wastes which could otherwise be returned to the land are not perceived Similarly the costs of consuming the forests are not incurred by the individual since the burden of replenishing the forests usually falls on the state Putting all these factors together it is not uncommon to find that social incentives to grow trees greatly exceed individual incentives in many areas and when properly accounted for in economic analysis will indicate that forestry activities are economically justified even though no single individual farmer will find it profitable to do so

The incentive to grow trees for woodfuel is obviously stronger where there is a commercial market offering financially attractive returns to tree growers This may be in local towns or more distant c1t1es However the returns to the farmer must generally not only be sufficient to justify his investments in wood production but greater than those from other potentially competing crops Where wood is grown on hilly lands farm borders etc that are not suitable for food crops the incentive to grow trees could be sufficient to make this effort worthwhile In these cases reductions in grazing land for animals or forage production as a result of tree growing may need to be considered carefully

When estimating these incentives it is essential to compare the prices received by the farmer and not final market prices Because of transport costs profit-taking by distributors and the costs of splitting firewood the producer may receive as little as 5-10--and

- 91 shy

exceptionally only 1--of the urban retail price For example in the early 1980s the ratio of the retail price in Blantyre (Malawi) to the typical rural producer price was around 201 [French 1985] and in Managua (Nicaragua) about 151 [Van Buren 1984] In Niger the license

Table 41 Potential Benefits of Rural Tree Growing

Benefit Type

Basic Resource Base Sol I protection Reduce wind and water erosion social

- sustain or enhance crop production private

Watershed protection Reduce siltation of upland rivers and regulate stream flows social - reduce frequency and severity of flooding - promote more even water flows reduce

irrigation requirements downstream - reduce siltation of irrigation and

hydropower systems

Agricultural Resources Moisture retention Preserve soil moisture (field trees) - Increase crop yieldsreduce irrigation needs private

Mineral nutrients Increase nutrient recycling and pumping from (field trees) deeper soil layers

Provide nitrogen with N-flxing species private Increase crop yieldsreduce needs for manure or chemical fertilizers

Forage from leaves increase animal production private - release crop residues and land for other social

uses than feed supply

Fruit nuts etc improve diet quantity and quality private income from sales

Timber - provide materials for construction basic private tools craftwork etc for local use income from sales

Windbreaks - reduce soil erosion shelter for animals social in extreme climatic conditions private

Energy and Other Woodfuels improve local householdartisanal supplies private

of firewood andor charcoal income from sales if commercial markets exist private and are profitable

Employment and development - provide employment broaden horizons and social range of activities increase participation in local decision-making etc IFAO 1978)

Ornament and shade - enhance environment social

- 92 shy

fee for cutting one stacked cubic meter of wood from the forest (stumpage fee) was recently about US8cent or less than 1 of the market selling price [Timberlake 1985] Transport and other marketing costs are discussed further in Section D

C FUELWOOD RESOURCES AND PRODUCTION

This section provides some basic data on and methodologies for assessing fuelwood supplies both from natural and managed resources It also discusses transport costs and other factors which play an important part in evaluating the economics of biomass fuels

Measurement Units and Concepts

Chapter I discussed the basic units for measuring the energy content of fuels and the moisture content density and volume of biomass fuels These concepts are not repeated here Basic data on the energy content of fuels are provided in Annex 1 For the biofuels these data should be used only for first cut estimates because of the substantial variation that is likely to occur with different tree species and moisture content levels

For estimating wood resources and actual or potential wood supplies one must first make a clear distinction between (1) standing stocks and (2) resource flows ie the rate of wood growth or yield Other important distinctions for energy assessments are

a Competing uses of the wood for timber construction poles etc These can be allowed for by estimating the fraction of the wood resource or yield that is available as a fuel resource under current conditions of collection or market costs and prices

b The fraction of the standing stock and yield that is accessible for exploitation due to physical economic or environmental reasons This quantity applies to natural forests and plantations for purposes such as watershed protection rather than to managed plantations village woodlots or single tree resources For example parts of a natural forestplantation may be on inaccessible hilly terrain or too remote for access except at prohibitive cost A study by FAO [de Montalembert and Clement 1983] estimated that physical accessibility of fuelwood from natural forests varied from 5-100 with 40-50 as a range that was often used in est ima tes Envi ronmenta 1 accessibility is often related to the minimum standing stock that can be left in situ without permanent degradation of soil or other resources

- 93 shy

c The fraction of the total yield that can be cut on a sustainable basis Total yield is usually referred to as the Mean Annual Increment (MAL) of stem wood normally in terms of solid volume per unit area (ie solid m3hectareyear) The sustainable yield might be lower than the MAL to protect the soil structure and nutrient recycling function served in part by dead and fallen wood in the soil

d The fraction of the cut wood that is actually recovered (harvested) ie allowing for collection and cutting losses which usually exceed 5 and may be much higher

Estimating Stock Inventories

The standing stock of trees is normally estimated by aerial surveys or satellite remote sensing to establish the areas of tree cover by categories such as closed forest open forest plantations and hedgerow trees etc Data must normally be checked by observations on the ground (llground truth) These observations are also needed to estimate tree volumes species type and perhaps growth rates (eg MAL) Inventory data is normally held by national Forest~ Departments and reported on a regional basis either as a volume (m ) in a given area or as a mean density (m3ha)

Inevitably estimates of tree stocks are approximate Furthermore most inventory data are for the commercial timber volumes which are a small proportion of total standing biomass The quality of fuelwood biomass may greatly exceed the commercial timber volume The most serious data deficiency in most countries is the lack of time series information to show where at what rate and due to what causes tree loss has been occurring

Estimating Supplies Stock and Yield Models

Incorporating the concepts outlined above Table 42 estimates the amount of wood that can be obtained from a natural forest by (1) depleting the stock and (2) by sustainable harvesting Essentially the method involves simple multiplication to adjust stock and yield quantities by the accessibility and loss factors mentioned above (Gowen 1985) The table also uses the concepts discussed in Chapter I to convert the volume yield of wood to an energy value

This model could apply equally well to a managed plantation or village woodlot although with different numbers to estimating the effects of forest clearance for agriculture (partial or complete stock loss) and to evaluating the impact of fuel gathering on forest stocks Furthermore the method is easily adapted to a time series model in which standing stocks are augmented (or depleted) each year by the difference between Mean Annual Increment and wood removals Finally the same model can be disaggregated to allow for different tree species and selective cutting methods Each major species will normally have

- 94 shy

Table 42 Example of Stock and Yield Estimation Method Natural ForestPlantation (Hypothetical Data)

Assumptions Stock Data Yield Data

Supply Factors

A Forest Area 1000 ha B Stock Density 200 m3ha

3C Stock Volume 200000 m

D Mean Increment 04 m3hayr

F Sustainable Yield 38 m3hayr3G Gross Sustainable Yield (A x F) 3800 m yr

H Fraction Available for Fuelwood 04 04

I bull Fraction Accessible 09 09 J HarvestCutting Fraction 09 09

K Gross Sustainable Harvest 3078 m3yr (G x I x J)

L Fuelwood Sustainable Harvest 1231 m3yr (K x H) 123 m3hayr

Clear Fell ing

3M Gross Harvest (C x I x J) 162000 m3N Fuelwood Harvest (M x H) 64800 m

O Wet Density (08 tonsm3)

P Net Heating Value (15 GJton or MJkg)

Q Energy Harvest Clear Fell ing 777 TJ ~ (N x 0 x P)

R Energy Harvest Sustainable 146 TJyr (L x 0 x P) 146 GJhayr

S Other Wood Clear Felling 77 700 tons (M - N) x 0

T Other Wood Sustainable Harvest 1477 ronsyr (K - L) x 0 147 tonshayr

a TJ = terajoule = 1000 GJ

- 95 shy

different stock volumes MAls and suitabilities for fuel or other wood resources In addition different cutting techniques for the same stock will imply different MAls

Estimating Financial Returns Plantation Models

When assessing the economics of managed plantations and wood lots normally one must estimate costs and benefits through time There are obvious analytical reasons why this is so for example to estimate annual cash flows compare net present values or rates of return on various projects or to estimate the loans andor subsidies needed to tide the producer over during the period between establishing the plantation and harvesting the first wood crop

There are two further reasons almost unique to tree growing why life cycle cost models are needed First with the exception of regular coppicing or pruning wood is harvested in different quantities at intervals of several years The supply is therefore lumpy and irregular and to provide a continual supply trees must be planted at phased intervals Second as trees mature and their diameter increases the value of wood also increases (in real terms) and may well exceed the value at which it would be sold as a fuel In other words while trimmings and thinnings at an early stage in the growth cycle (rotation) may be used locally or sold as woodfuel at later stages-shyand especially after the final clear felling--much of the wood will probably be used or sold as timber and not fuel

Table 43 provides an illustration of a life cycle cost analysis in which annual costs and benefits are recorded from plantation establishment to final felling on a 20-year cycle It is based on Pakistan Forestry Department data for plantations of shisham trees for timber and fuelwood Returns from forage leaves and other byproducts are ignored The method can easily be adapted to rotations of any length and to the assumption of constant wood prices (in real terms)

- 96 shy

Table 43 Example of Financial Discounted Cash Flow Method Plantation (Data Based on Irrigated Shlsham Plantation Pakistan)

Per Hectare Costs Per Hectare Production Cash Non-harvest Harvest Volume Value Revenue Flow

Year ($) ($) (m3) (Im3) ($) ($)

1 330 - 330 2 165 - 165 3 130 - 130 4-5 60 - 60 6 60 37 209 353 738 + 641 7-10 60 - 60

11 60 81 456 530 2417 +2276 12-15 60 - 60 16 60 73 343 706 2422 +2289 17-19 60 - 60 20 60 375 1515 882 13362 +12927

TOTALS 1645 566 2523 18939 +16728

Net Present Value (10 interest) a + 3037 (Costs amprevenues fa 1 In mid-year)

General data

454 ha irrigated plantation initial spacing 3 x 2 m (1793 seedlingsha) land rent of $75ha excluded Costs converted from Rupees at Rs 10$

Cost data per hectare

All years irrigation $30 maintenance (including watercourses) $30 Year 1 establish plantation (site preparation layout digging water

channels plant costs plant transportation planting) S200 ~ restocking $35 Years 1-3 weed Ing $70

Harvest data and costs

Year 6 1st thinning at SI77m3

Year II 2nd thinning at SI771m3

Year 16 3rd thinning at S2121m3

Year 20 final felling at S247m3

~I NPV ca I cu Iat Ion For each year net costs or revenues are mu I tip lied by a discount factor For a 10 discount rate and mid-year costs amp revenuesthe factor is 111

raised to the power of (N - 05) where N is the Year Number The annual values are then summed

~ PFI (1981)

- 97 -

Fuelwood Production Data

Table 44 provides data on typical fuelwood tree species by clim~tic zone The table also gives the basic densities of the woods in kgm since these densities are needed to convert volumes to weights In general densities are lowest (400-600 kgm3 ) for young trees a~d for fast-growing species They may be much lower still (200-400 kgm ) for eucalyptus and other fast-growing fuelwood species on very short 1-3 year rotations since the harvest is mostly in the form of small branches twigs or shoots and leaves In contrast mature trees of slow-growing species have much higher densities in the 500-1000 kgm3 range

Table 44 Characteristics of Various Fuelwood Species

Fuel wood Average Average Basic Species Rotation Production Density

(yrs) (m3hayr)

Humid Tropics Acacia a

aurlc- I I form s good soil s

poor sol Is Cal I iandra calothyrsus ~

1st year 2nd year

Casuarlna b equisetlfolla

Leucaena b leucocephala

Sesbanla blspinosa S grandlflora

Tropical Highlands Eucalyptus globulus E grandis irrigated

Good sol Is Poor sol Is

AridSemi-Arid Acacia sallgna A Senegal

Gum plantations Wood plantations

Albizia lebbek a Azadiarachta indica a Cassia slamea Eucalyptus

camaldulensis good sol Is poor sol Is

E citriodesra ~I

Prosopls jutiflora good sol Is poor soi Is

10 - 12 4 - 8

7 - 10

8 - 10 6 ms 2 - 5

5 - 15 5 - 10 5 - 10

10

4 - 5

25 - 30 15 - 20 10 - 15 8 5 - 7

7 - 10 14 - 15 8

10 15

17 - 20 10 - 15

5 - 20 35 - 60

10 - 20

25 - 60 15 odthayr 20 - 25

10 - 30 40

17 - 45 5 - 7

15 - 10

05 - 10 5 - 10 5

10

10 - 15

20 - 30 2 - 11

15

7 - 10 5 - 6

06 - 08 06 - 08

05 - 08 05 - 08

08 - 12

03 04

08 - 10 04 - 05 04 - 05 04 - 05

(lIght)

(heavy) (heavy) 05 - 060 06 - 09 06 - 08

06 06 08 - 11

07-10 07-10

al Preferred fuel wood speciesbl Preferred fuel wood and charcoal species

Source NAS [19801

- 98 -

Fuelwood Market Prices

Fuelwood prices are generally reported as retail or wholesale market prices usually for urban locations These are important to fuelwood users and producers but they largely ignore the benefits of tree cover (and costs of forest depletion) which include protection from soil erosion watershed protection and avoided costs of afforestation Economic prices therefore should be used in project analysis (See Section C for discussion of methodology)

Table 45 presents urban retail fuelwood prices in several developing countries As one might expect they vary widely from $10-140ton across countries and by as much as 31 within some countries The inter-country variation is partly explained by the use of market exchange rates to convert local currencies to dollars The rest of the variance is explained by (1) the cost of competing fuels I (2) the cost of transport and fuelwood preparation (eg splitting logs into firewood pieces) (3) quantities purchased (small bundles normally cost more per kg than bulk purchases) (4) quality (species size and size uniformity of split pieces) (5) locale within the city and (6) the sale value by producers The final item includes producer profit and the costs of producing and harvesting the wood resource The (marketgt production cost may be very small or zero when wood comes from land cleared illegally

for agriculture or or with a permit

is taken from public forests whether

Fuelwood Relative Prices

In some countries firewood and charcoal prices have been rlslng rapidly both in real terms and relative to alternative fuels such as kerosene and LPG In others they have fallen in real terms and have become progressively cheaper than fossil cooking fuels The addition or removal of subsidies particularly on kerosene complicates these relative prices Nevertheless in some places woodfuels are becoming so costly that there are strong incentives for consumers to switch away from them for cooking In these cases one needs to examine carefully the assumptions about projected demand on which woodfuel supply projects are based

The wide range in relative prices is indicated by data from 17 countries which show that the ratio of kerosene to firewood prices (per unit of delivered energy) varied from 03 in parts of Nigeria to 16 in a rural area in South Africa between 1980 and 1983 The ratio of charcoal to firewood prices varied much less as one would expect with the lowest ratio at 111 (Bangalore India) and the highest at 301 (Freetown Sierra Leone)

11 There is some evidence that in several countries woodfuel prices have risen in line with jumps in the prices of kerosene the main competitor to woodfuels

- 99 shy

Table 45 Retail Fuelwood Prices in Various Developing Countries

Cost of Cost of delivered utilized energy a energy b

RegionCOuntry Year USSton fMJ - fMJ - Source

Africa ---rtiiTop I a 1983 80-90 052 - 058 40-45 b

Gallbla 1982 140 090 69 b Gallbia (Banjul) 1982 53 034 26 a Kenya 1981 10 006 046 b Liberia 1984 50 - 130 032 - 084 25 - 65 b Madagascar 1985 20 - 25 013 - 016 10 - 12 b Malawi (Blantyre) 1981 37 024 18 a Morocco 1983 20 - 60 013 - 039 10 - 30 b Niger 1982 60 039 30 b Sudan (Khartoum) 1982 72 046 35 a

Asia --eangladesh (Dacca) 1982 38 025 19 a

BUnDa (Rangoon) 1982 60 039 30 a India (Bombay) 1982 87 056 43 a Nepal 1981 20-60 013 - 039 10 - 30 b Pakistan (Karachi) 1982 20 - 40 013 - 026 10 - 20 b Sri Lanka (Colombo) 1982 61 039 30 a Thai land 1984 17 011 085 a

Latin America Guatemala 1982 34 022 17 a

(Guatemala City) Peru 1983 20-60 013 - 039 10-30 b

Note Prices vary considerably by quantity purchased ~ Cost of delivered energy assumes heating value of 15500 MJton b Cost of utilized energy assumes end-use efficiency of 13J

Sources a FAO [1983a) b UNOPlWorld B

Bank ank Energy Sector Assessment Reports Washington DC The World

Normally relative prices are compared for utilized energy (sometimes called the effectivetl price) since this is the relevant measure for the consumer and for questions of fuel substitution a switch in fuel normally requires a corresponding switch in cooking appliance end-use efficiency and effective price The latter is calculated simply by dividing the delivered energy price (eg in $MJ) by the end-use efficiency of the appropriate end-use appliance Appliance costs (amortized so that they can be added to fuel costs) are frequently included in these comparisons

- 100 shy

Table 46 Relative Costs of Cooking In African Countries 1982-83

Cameroon Senegal NNigeria Niger Ethiopia

Relative Costs ~ Fuel wood 10 10 10 10 10

Charcoal 34 09 24 14 16

Kerosene 100 17 06 17 07 n8 13 - 19 20 20 1 bull 1 LPG

Electricity 111 33 11 28 20

Fuelwood Costs Cents per MJ of

nut iii zed heat b 1 bull 1 25 31 25 72

a Assuming thermal efficiencies of 13 and 22 respectively for cooking with fuelwood and charcoal using metal pots The fuelwood prices used in the calculations correspond to those found in urban centers and Include the costs of appliances

b That is per MJ of heat output by the stove and absorbed by the pot The nature of the trial on which the data are based is not described in some sources so it is not possible to provide a confidence interval for the estimates

Source Anderson amp F I shw ick [19841 us i ng data from UIf)PWor I d Bank Energy Assessment Reports

Table 46 compares the effective (utilized energy) costs of cooking with fuelwood charcoal kerosene LPG and electricity including equipment costs in five African countries in the 1982-83 period While in Cameroon woodfuels are the cheapest option in Ethiopia cooking with woodfuel is as expensive or more expensive than using most of the modern fuels

Table 47 presents a more detailed analysis of cooking fuel prices in Nigeria in order to show the methodology applied According to this table wood and charcoal are much more expensive than kerosene LPG or electricity for cooking even though LPG and kerosene are often difficult to obtain

- 101 shy

Table 47 Comparative Prices of Household Cooking Fuels in Nigeria

Fuel

(I)

Del ivered Price

(kunit)

(2)

Net HV (MJunit)

(3) End-Use Eff iciency

()

(4) Effective Price

(kMJ uti I ized)

Appl lance Cost

(N=IOOk)

Wood (air dried) Charcoal Kerosene

LPG Electricity

17kg 22kg lOll 281 34kg 6kWh

1471kg 251kg 3481 3481 490kg 36kWh

8-13 20-25 30-40 30-40 45-55 60-70

89 -44 -07 -02 -13 -24 -

145 58 10 27 15 27

na na 3 al

38 bl 40 45 40

Effective price (Col 4) = (Col 1)

(Col 2) x (Col 3)100

al Small one burner wick stove bl Two burner pumped stove N = Naira k = kobo (1 Naira = 100 kobo) Source UNDPWorld Bank [1983c]

Fuelwood Economic Values

Several methods have been used to depict the economic [social] value of fuelwood production in contrast to market (financial) costs and returns This can be done whether or not fuels have a commercial market price by establishing proxy values which reflect either the economic costs of alternative fuels that would be used if the fuelwood was not produced or the total benefits and avoided costs of tree planting It is important to note that the market prices are usually a poor guide to economic values in general they are likely to be much lower than economic values owing to the divergence between the individual and social costs of fuelwood cutting discussed before Also while there are several methods of calculating economic values limited data and other uncertainties usually make this task very difficult

Nevertheless one method of calculating economic values for fuelwood is to evaluate the opportunity cost of using the alternative fuel most likely to be used if wood were not available eg kerosene or crop residues and animal dung With residues or dung the method could involve estimating the economic cost due to the increase in soil erosion or loss in crop production that results from diverting the material to energy uses For example in a World B~nkFAO community forestry appraisal in Nepal it was estimated that 1 m of air-dried fuelwood was equivalent in energy terms to 568 tons of wet animal manure and that if the latter was used as manure rather than being burned it would increase maize yields by about 160 kghayr Given the market price of mai~e the economic value of fuelwood was estimated at Nepal Rupees 520m [SAR 1980]

- 102 shy

A second method is to evaluate the non-wood benefits such as savings in fuelwood collection time fodder values in terms of increased milk yields and their prices the value of shelterbelts in increasing crop yields or benefits in preventing soil erosion and desertification For example the same Nepal appraisal estimated the value of fodder using the following methodology (1) calculate the net quantity of leaf fodder and grass produced (2) from this estimate the fraction that would be fed to animals (3) estimate the increased milk yield due to this additional feeding and (4) calculate the value of the additional milk produced Over the 30-year project life the value of the leaf fodder was estimated to be US$11 million

Plantation Costs

The cost of establishing fuelwood plantations varies considerably depending on the terrain and amount of land preparation needed irrigation works (if any) labor costs and the like Table 48 presents data on 12 fuelwood projects financed by the World Bank during the early 1980s The range of investment costs varies from US$212ha to 2000ha (1984 dollars) although there are substantial economies of scale associated with plantation area If the two projects of 5000 hectares and below are excluded the range narrows to $212-934ha

Smaller scale social and community forestry schemes should cost less than fuelwood plantations since much of the labor is provided by the recipients of the scheme In the Karnataka Social Forestry Project India plantation costs ranged from only US$51ha for bamboo in tribal areas to US$464 for plantings on public waste lands (1983 dollars) Administrative and equipment overheads for the whole scheme ignoring contingency estimates averaged about $lOOha [SAR 1983]

Apart from initial investments the important cost with plantations is the final harvest cost per unit of wood This varies widely by climate species irrigation and other input costs--and above all tree survival rates The cost of harvesting and transport generally amounts to $ 15-20m3--at least twice that of establishment Most available sample figures are based on pre-project estimates and therefore may bear little relation to actual results Suffice it to say that some appraisals have suggested that plantation fuelwood can be produced at less than current market prices and with even lower economic costs As a general rule these tend to include a high level of participation by local people In contrast large scale plantations in unfavorable climatic zones can prove to be prohibitively costly For example World Bank assessments of fuel wood planttions in the arid regions of Northern Nigeria gave costs of US$74-108m By comparison the price at which fuelwood delivered to urban ~rkets became uncompetitive against kerosene and LPG was about US$70m bull

Table 48 Selected Fuelwood Projects Financed by the World Bank Since 1980

Year of Approximate Loan or Afforestation End Products Other Investment

Country and Project Credit Area Main Species Than Fuelwood al Cost per ha (ha)

=

1984 US$ I

Upper Volta Forestry 1980 3500 Euc Gmel ina Saw logs 1867 pound1 India Gujarat 1980 205000 Alblzla Acacia Poles 672

bamboo Casuarlna Prosopls Morus

Malawi NRDP IIWood Energy 1980 28000 Euc Glnel ina 467 Nepal Community Forestry 1980 11000 Alnus Prunus Fodder poles 840

Betula Pinus Rwanda Integrated Forestry amp Land 1980 8000 Euc pine Saw logs 934 Bangladesh Mangrove Afforestation 1980 40000 Mangrove spp Pulpwood saw logs 373 Tha I I and Northern Agriculture 1980 11000 Euc pine Poles 212 Senegal Forestry 1981 5000 Euc neem Poles 2000 India West Bengal 1982 93000 Euc indig spP Poles fodder fruit 312

0 bamboo w

Niger Forestry II 1982 8650 Euc Ac neem Poles 784 India Jammnu Kashmir Haryana 1983 111500 May Incl Indig Small timber 502 Zimbabwe Rural Afforestation 1983 5200 To be determined Poles 616

Unweighted mean 798 Weighted mean 559

In this column poles refers to building poles mainly for traditional construction ~ The US$ amounts were converted from current to 1984 values by means of the Manufacturing Unit Value (MUV) Index which is published

per I od I ca II y by the Econom i c Ana Iys I s and Project ions Department of the Wor I d Bank th i s Index ref Iects both Internat i ona I Inflation and changes in the US$ exchange rate and the latter changes in turn reflect (Ia) differences between local and US inflation rates The investment costs include not only the immediate afforestation costs including weeding and after-care until the trees are firmly establ ished but also some related investments in studies training and Institution-building They also include physical contingencies

pound The often very high cost of afforestation in the Sahel countries is generally due to a combination of difficult ecological conditions and overvalued exchange rates

- 104 shy

D TRANSPORT COSTS AND HARKET STRUCTURES

Urban woodfuels are sometimes trucked or brought by rail over long distances Transport costs thus may be a critical component not only of urban woodfuel prices but of the area from which woodfuels can be supplied at competitive prices Potential resources which are otherwise economically attractive may be ruled out due to transport distances and costs thus limiting supply possibilities as urban demands for woodfuels expand unless fuel prices incre~se substantially Because fuels with the highest energy densities (MJm or MJkg) are the cheapest to carry transport costs (other factors being equal) reduce the relative prices-shyand increase the availability--of urban fuels such as charcoal and densified biomass compared to firewood

Examples of transport costs and their impact on retail prices are presented below and examples comparing costs and maximum economic transport distances for firewood and charcoal are provided in Table 49 Before turning to these some general points about transport costs may be in order

a Transport costs are often quoted per ton-kilometer But stacked firewood and to a larger extent charcoal have such low densi ties that the load which a truck can carry may be limited by volume and not weight

b In many areas (eg the Sahel) woodfuel is trucked by small informal owner-operators in 15-20 year old vehicles which have very low overhead costs such as depreciation maintenance spares and insurance Their costs may be one third to one half of those charged by large commercial enterprises For example in Nigeria about 65 of trucking costs are attributed to depreciation maintenance spare parts and overheads 14 to wages 10 to tires and only 11 to fuel and lubricants [FMT 1983]

c Woodfuels are sometimes carried as partial loads and on empty return trips and so have very low or zero opportunity cost This applies especially to small urban markets in parts of Africa

These factors help to explain the considerable variance in fuelwood transport costs that have been found in surveys The results of several World Bank [Schramm amp Jirhad 1984] assessments and those done by others illustrate this point

In Zaire woodfue1 transport costs US$011-024 per ton-km over unpaved roads but only US$07-14 per ton-km over paved roads

- 105 shy

In Nigeria (1983) firewood transport in 10-ton trucks typically costs only US$055 per ton-kin but for comparative short trips of 100 kin can account for as much as 50 of the ex-woodlot price

In Ghana (1980) charcoal transport costs were much lower still at US$0065 per ton-km for the 350-kIn trip from Accra to Nima Nevertheless transport accounted for about 50 of the wholesale market price [Schramm amp Jhirad 1984]

In Ethiopia (1983) the financial costs of carrying briquet ted cotton residues in 22-ton trucks over 300 km were estimated at US$14ton plus US$2ton for handling charges glvlng a total transport charge (less bagging at US$38ton) of US$024ton-km This was 36 of the delivered cost to the urban market [Newcombe 1985] bull

In Nicaragua (1981) fuel wood transport in 5-ton trucks cost about US$Olton-km for the 150 kin trip to Managua where it accounted for 27 of the retail price [Van Buren 1984]

Table 49 provides a formula for estimating woodfuel transport costs It shows that for any but the shortest trips when handling charges are significant costs are inversely proportional to the load and the energy density of the fuel (GJton) Since charcoal has roughly twice the energy content per unit weight (MJkg) of firewood it costs approximately half as much to carry Costs are also directly proportional to the load carried and cost per vehicle-km as one would expect

Table 49 also gives an example comparing the maximum transport distance for firewood and charcoal using hypothetical but realistic values This shows that the maximum distance is extremely sensitive to the difference between the Itproducer pricelt

- (at the point of loading) and the maximum Itdelivered price at the market (the price at which the fuel remains competitive) Some fixed costs such as for bagging charcoal and splitting firewood have been ignored although they obviously affect the producer and delivered prices The delivered price of charcoal has been set at just over twice the firewood price to allow for its greater end-use efficiency

The example shows that (with these data) the maximum distances for firewood and charcoal are about 170 km and 990 km respectively a ratio of roughly 1 6 However the area from which fuels can be transported competitively is in the ratio of 136 This example helps to explain why charcoal is sometimes trucked over distances of 600-900 km to urban centers and can lead to tree loss over vast areas It also emphasizes the importance of drying biofuels before transport and densifying them to briquettes or pellets if this is logistically possible

- 106 shy

Table 49 Woodfuel Transport Costs General Formula and Example

General Formula for a Single Trip (weight basis)

F I loadingunloading cost (fixed cost) May be calculated from load (tons) x costton l tons Weight of load carried (assumed all woodfuel) C Ilkm Trucking cost per vehicle - km T k Trip length E GJton Energy density of fuel as transported P IGJ Cost or price to point of loading (producer energy price) May be calculated from

other units such as Iton and GJton 0 $GJ Cost or price at point of del Ivery (dellvered energy price)

Note 0 = P + transport cost in IGJ

Trip cost F + CxT Trip costton load (F + C x nil Trip costGJ (F + C x T)(l x E)

To estimate the maximum competitive trip length (Tmax) we can set the del ivered energy price to a maximum value that the market will bear (Omax) Then

P + (F + C x Tmax)(L x E) lt Omax which gives

Tmax lt (Omax - P) x L x E - FC

(Volume basis) If the load Is limited by maximum volume rather than weight the values land E can be converted to volume units (m3 GJm3) Note that stacked or packed volumes and not solid volumes must be used

Worked Example for Firewood and Charcoal

Basic parameters Firewood Charcoal Both

Producer price $1m3 20 40 Bulk density tonsm3 06 025 Producer price Ston 333 160 Energy content GJton E 155 300 Producer price SGJ P 215 533 Del ivered price SGJ (max) 0 30 70 load tons l 10 Loadunload cost$ F 10 Trucking cost Svehlcle-km C 1 Applying the formula for max distance Max trip length for given conditions km 168 989 Supply area km2 89000 3072000

- 107 shy

The difference in supply area can be very much greater than this In some parts of Africa charcoal can be transported economically over a direct distance of 600 km giving a potential (under straight road conditions) concentric supply area of up to 11 million km2 (110 million hal around a city Even with a mean annual yield from farm and forest areas of only 025 m3hayr this area would yield 28 million m3 of fue1wood annually enough to supply around 25-30 million people Assuming that in the same area firewood can be economically transported over a direct distance of 70-100 km--as estimated in some World Bank assessments--the firewood supply area would be only 1 of the charcoal supply area

E CHARCOAL

In many cities of Africa and Asia charcoal is fast becoming the dominant fuel where wood resources are scarce or located far from urban centers One major reason for this trend is the lower transport cost and greater supply area of charcoal as outlined above Other advantages are that charcoal is easier for the consumer to carry from the market due to its greater energy density (MJkg) is easier to handle and store gives a more even cooking temperature than wood and with suitable equipment has a higher end-use efficiency Also charcoal is smokeless and can be used indoors offering greater convenience This is especially favorable in urban areas For many consumers these advantages outweigh the fact that (typically) it costs more per kg than firewood However charcoal may require more wood resources than the direct burning of fuelwood A good recent review of charcoal issues appears in Foley [1986]

Production Processes and Yields

Charcoal can be produced in batch or continuous kilns retorts or furnaces but the basic principles are the same for all technologies Combustion is initiated in a wood pile within the conversion device and proceeds with a very limited supply of air until the wood is reduced to charcoal This process is often called carbonization

Most charcoal is made from wood although other sources may include coconut shell coffee husks (eg Ethiopia) cotton stalks (eg Sudan) and timber wastes Excess bark in the wood results in charcoal that is friable and dusty However charcoal fines dust and small fragments can be briquetted The type of equipment density and moisture content of wood govern the charcoal yields from a kiln or retort Dry and dense wood yield the highest proportion of charcoal as a percentage of the orginal wood weight (oven dry) (See Table 410 below) Yields also tend to be greater with larger kiln size and also depend on the amount of charcoal dust or fines produced Fines arise both in the charcoaling process and from vibration and shaking of finished charcoal pieces during handling bagging and transport Up to 30 of charcoal may

- 108 shy

be fines on removal from the kilnretort although fines typically are much less than this a further -2~Q of lump charcoal may be broken down to fines during transport over poor roads Bagged charcoal in the market may contain from 5-20 fines Although fines can be briquet ted and sold often simply by hana-tosses and increased unit costs are inevitable

The effects of wood density moisture content and conversion technology on charcoal yields are shown in Table 410 adapted from Openshaw [1983] Apart from inherent differences in conversion technology th~ effects of greater density and the use of drier wood on charcoal yields are clear If one includes the technological variations the complete range of yields (and energy conversion efficienciesgt is a factor of six to one

Table 410 Yields and Conversion Factors for Charcoal Produced from Wood

Effect Of Wood DensitySpecies Average Preferred Mangrove

Pines Tropical Hardwood Tropical Hardwoods (Rhizophora)

Charcoal yields

kg per m3 wood 13 moisture wet basis 115 170 180 185

kg per m3 wood oven dry basis 132 195 207 327

Effects of Technology and Moisture Content

For typical preferred tropical hardwoods

Oven dry weight of wood (tons) to produce one ton of charcoal including fines (approximate data)

Moisture dry basis 15 20 40 60 80 100 wet basis 13 167 286 375 444 50

Kiln type Earth ki In 62 81 99 130 149 168 Portable steel ki In 37 44 56 81 93 99 Brick ki In 37 39 44 62 68 75 Retort 28 29 31 44 50 56

Energy Conversion Efficiency percent ~

25 ~~Earth ki In 19 16 12 10 9 Portable steel kifn 43 36 28 19 17 16 Brick ki fn 43 40 36 25 23 21 Retort 56 54 51 36 32 28

~ Assuming wood at 20 MJlkg oven dry charcoal at 315 MJlkg 5 moisture (wet basis) including fines

Source Adapted from Openshaw 19831

- 109 shy

This brings us to the much-debated point whether charcoal is more wasteful of wood resources for cooking than direct wood burning Many authors have asserted that it is and they are obviously correct if one assumes that charcoal is made from wet green wood in primitive earth kilns where the wood-charcoal conversion efficiency is only about 9-12 in terms of energy as opposed to weight (See Table 410) The greater energy efficiency of cooking by charcoal rather than wood fires or stoves cannot generally make up for this difference However as shown in Table 35 of Chapter III end-use efficiency of a metal charcoal stove with aluminium cooking pots is 20-35 and that of an open fire with clay pots is about 5-10 or 35-4 times less Thus if consumers switch from an open wood fire using clay pots to a charcoal stove with aluminium pots and wood-charcoal conversion efficiencies are better than 25-28 wood consumption will fall when charcoal is used instead of firewood This efficiency rate or better is achieved with all the technologies except for earth kilns as long as fairly dry wood is used

Nevertheless these arguments underline the importance of using high quality data preferably from large sample surveys in carrying out any assessment of woodfue1 resources charcoal conversion technologies and cooking fueldevice substitutions Sensitivity analyses should also be made to check the effects of errors in the basic data and it should be recognized that this is one area of energy analysis where rules of thumb are frequently inaccurate

Charcoal Prices and Other Data

Since charcoal is almost pure carbon its heating value varies little by wood species Gross heating values oven dry are about 32-34 MJkg When air dried the moisture content (wet basis) is typically about 5 and the net heating value is close to 30 MJkg In damp weather charcoal easily absorbs water and its moisture content may rise to 10-15 For this reason lower net heating values of about 27 MJkg are often reported in the literature

Table 411 provides a list of wood characteristics and their advantages and disadvantages for charcoal making Just as there are strong preferences for types of firewood so too with charcoal Many consumers are very selective about its hardness friability density the size of pieces and burning quality

- 110 shy

Table 411 Preferred Wood Feedstock Characteristics for Charcoal Production

Wood Characteristics Reason

Mature Tree not too young or too 0 I d

Thin Bark

Compact Heavy

Correct Dimensions

Healthy

Low Mol sture

Very young trees are rich in sap and thus have high moisture content trees that are too old have longitudinal fibers that separate creating a friable charcoal product or fines

Bark can be very rich in ash which makes a poor quality charcoal

Light or loose woods often result In charcoal with low compressive strength so that it breaks easily and produces fines

Wood that is too thick (diameters over 25 cm) (length diameter) or too long (longer than 180 or 200 m) slows down the carbonization process leaving semi-carbonized pieces of wood In the final product

Wood that has been attacked by fungus or other depredations gives lower yields It also makes low quality charcoal which Is friable and fragi Ie

Moisture levels above 15~ to 20~ slow the carbonization process and lower the conversion efficiency

Source Osse (1974)

Table 412 shows retail charcoal prices in a number of countries Once again the ranges are large and are explained by factors similar to those for wood prices producer and transport costs wholesale versus retail costs charcoal quality and the size of the sacks or bags in which charcoal is sold Typically charcoal production costs account for 50-65 of the retail price while transport makes up 15-30 of the final price [UNDPWorld Bank 1984c] For simple charcoal production technologies such as earth kilns the wood feedstock cost dominates the costs of production though the significance of feedstock costs in financial terms depends greatly on whether wood is purchased or freely collected

- 111 shy

Table 412 Retai I Prices of Charcoal In Selected Developing Countries (per 30 kg bag sold at markets)

Cost of Cost of Reglonl Charcoal Net Heating Del ivered Systetll Uti I Ized Country Price Value Energy ~ Eff iclency Energy ~I

($kg) ~ (MJkg) (fIMJ) () (fIMJ)

Africa Ethiopia ( 1983) 044 29 07-1 7 23 30 - 74 Kenya (1981) 006 29 02 23 09 Li ber i a (1984) 014 - 022 29 05 - 08 23 22 - 35 Madagascar (1984) 009 - 017 29 03 23 13 Niger ( 1982) 015 29 05 23 22

Asia Thai land (1984) 009 - 021 29 03 - 07 23 13 - 30

Latin America Peru (1983) 038 29 13 23 57

al Cost of delivered energy aSSUMeS a heating value of 29 MJlkg at 5 mcwb bl Cost of utilized energy aSSUMeS an end use efficiency of 23bullbull equivalent to most

efficient traditional charcoal stoves as measured in World Bank sector work in Ethiopia and Liberia Efficiency range is 15 - 23 for traditional and 25 - 40 for improved stoves

cl Converted at Official exchange rate

Sources UNDPlWorld Bank Energy Sector Assessment Reports

F AGRICULTURAL RESIDUES

In wood-scarce areas raw agricultural residues are often the major cooking fuels for rural households The greatest concentration of residue burning is in the densely populated plains of Northern India Pakistan Bangladesh and China where they may provide as much as 90 of household energy in many villages and a substantial portion in urban areas too For many people in these areas--some of which were deforested centuries ago--the woodfuel crisis is essentially over The evolution of fuel scarcity has entered a new phase where the struggle is not to find wood but to obtain enough st raws (andmiddotmiddot animal dung) to burn [Barnard amp Kristoffersen 1985] while knowingly risking the threats of--or causing--soil erosion nutrient loss and reduced agricultural productivity that result from excessive residue removal Hughart [1979] has estimated that 800 million people now rely on residues or animal dung as fuel although reliable figures are scarce

- 112 shy

Residue Supplies and Energy Content

Most farming systems produce large amounts of residues With most cereal crops at least 15 tons of straws and husks are produced for each ton of grain [Newcombe 1985] With other crops such as cotton pigeon pea and coconuts the residue to crop ratio can be as high as 5 1 This means that in the rural areas of many countries average residue production exceeds one ton per person [Barnard amp Kristoffersen 1985] Table 413 provides some data on residue to crop ratios and Table 414 gives heating values for some major types of residue

Table 413 Residue-to-Crop Ratios for Selected Crops

Residue Production Crop Residue (tonnes per tonne of crop)

Rice straw 11 - 29 Deep water rice straw 143 Wheat straw 10 - 18 Maize stalk + cob 12 - 25 Gra I n sorghum stalk 09 - 49 M Ilet stalk 20 Barley straw 15 - 18 Rye straw 18 - 20 Oats straw 18 Groundnuts shell 05

straw 23 Pigeon Pea stalk 50 Cotton stalk 35 - 50 Jute sticks 20 coconut (copra) shell 07 - 11

husk 16 - 45

Source Barnard ampKristofterson [19851 See also Newcombe (19851

- 113 shy

Table 414 Calorific Values of Selected Agricultural Residues (MJkg oven dry weight)

Ash Gross Heating Value Material Source Content (oven dry weight)

Alfalfa straw

Almond shell Cassava stem Coconut she I I Coconut husk Cotton stalks

Groundnut shells

Maize stalks

Maize cobs

01 ive pits Pigeon pea stalks Rice straw

Rice husks

Soybean stalks Sunflower straw Walnut shells Wheat straw

(1 )

(1)

(2) (3)

(3)

(1) (4) ( 1 )

(4) (1)

(4) ( 1 )

(4)

(5) (4)

(5) (4) (1)

(2) (1)

(I)

(1)

( 4)

()

48

08 60

172 33

44 64 34 15 18 32 20J

192

165 149

11

85

(MJkg)

184 173 194 183 201 181 158 174 197 200 182 167 189 17 4

214 186 152 150 153 155 168 194 210 211 189 17 2

Sources (l) Kaupp and Goss 119811 (2) Saunier et al 119831 (3) KJellstrom [19801 (4) Pathak and Jain 119841 and (5) OTA 11980)

Viewed purely as a fuel residues can be a large resource However as discussed in Section B most residues have important or vital alternative uses quite apart from the need to leave some of them in the field to retain moisture reduce soil erosion by wind and rain maintain or enhance soil nutrients and preserve the physical structure of the soil Their use as fuel has to compete with these alternatives although in many places the cooking fire has to take precedence The supply of crop residues for fuel can be estimated by a formula which allows for these alternative uses and is based on a method [Gowen 1985J very similar to the one used in Table 42 to determine wood yields from forests

- 114 shy

(1) (2) (3) (4) (5) Potential Crop Crop Residue Fraction Fraction Residue = Area x Yield x to Crop x ava flabl e x avai lable Supply Ratio allowing for susshy allowing for

talned soil fertility non-energy uses (tyr) (ha) (thayr) (xix) (xix) (xix)

Items 4 and 5 can be expressed as weights and subtracted from the product of Items 1 2 and 3

Given the large range of residue to crop ratios--varying significantly within the same crop species by cultivar--and crop yields there is little point in providing typical figures of residue production per hectare or the availability of this residue as fuel Local data on residue availability must be used instead

With residue analysis a clear distinction must be made between (1) material that is left in the field after harvesting but which can be collected later (eg wheat straws and stubble) and (2) crop husks and shells that are harvested with the main crop product and separated during processing (eg rice and coffee bean husks wheat chaff coconut husks and fiber) Collection costs for the first type are often prohibitive With the second type residues are frequently collected with the main crop product and brought to a central processing point

A further distinction must be made between distributed and concentrated collection due to the differences in volumes flowing into the collection point Distributed production refers mostly to familyshyscale crop processing which produces small volume flows at a multiplicity of locations Residues may be used by the family or in the village but the costs of transporting them to a central depot for further processing are likely to be prohibitively high Moreover these small farm residues often have higher value uses as animal feed roughage and soil conditioner Concentrated production produces large volumes at just a few locations Examples are the processing plant of a large cash crop farm a village rice de-husking plant and sawmill wastes In these conditions it may well be economic to process residues into briquettes or pellets or convert them to other forms of energy such as biogas producer gas or electricity via the boiler and steam cycle

Availability and Economic Costs

A central question emerges whenever crop residues and animal wastes are considered as possible fuel sources How much safely can be harvested The question is the source of vitriolic argument and a large literature reinforced by data that is confusing conflicting or absent entirely This section will not attempt to resolve this dispute but instead will provide some guidelines to the main issues

In some arid and semi-arid areas where biological productivity is already low there is no question that after the trees have been

- 115 shy

cleared and people have begun to burn residues and dung from the fields in large quantities severe soil degradation and reductions of crop yields begin As productivity falls and local people press harder on the remaining resources the biological system can slide down into a terminal stage of almost total collapse This transition is occurring across Ethiopia and in some areas has reached the terminal phase although the burning of crop residues may not be the sole cause of this collapse The same transition can be seen in other parts of Africa A graphic account of the stages of this transition is included in Annex 9 taken from Newcombe [1984b]

At the opposite extreme it has been argued that in moist temperate zones all residues can be removed from the field without any serious effects on soil health provided sound agronomic practices are followed [Ho 1983] including crop rotations and sequencing strip cropping contouring or terracing and use of chemical fertilizers Much of the required organic matter is provided by the sub-surface root systems of crop plants which are not considered here as removable residues

There are three main issues involved in removing residues from tropical and semi-tropical farming systems

Depleting Organic Matter Under steady state conditions additions and losses of organic matter in the soil are in approximate equilibrium If less residue or dung is returned to the soil the organic matter content will decline slowly until a new equilibrium is reached However there are virtually no data on tropical farming systems to establish the rate of decline or how far it will go under different crop and management conditions [Barnard amp Kristofferson 1985] Losses of 30-60 over a few years have been recorded when forest land is converted to agriculture but this has little relevance to land under continual farming

Reduced Nutrient Balances The effects on crop productivity vary greatly according to the crop and farming system With low input dryland agriculture as in the poorest parts of the developing world chemical fertilizer use is low and organic matter breakdown is the principal source of nitrogen and sulphur and a major source of phosphorous If reserves of these nutrients fall sufficiently crop yields will be reduced--although the degree and rate of reduction depend on many factors including the initial nutrient levels and the amount of nitrogen fixing by plants (eg legumes and some tree species) With low input wetland or irrigated farming (eg rice cultures) significant amounts of nutrient are provided by the irrigation water and nitrogen fixing organisms Even substantial reductions in organic matter levels may be possible without serious effects on crop yields

In wet and rainfed systems the enormous range of effects is well illustrated by the results of l2-year trials to increase residue

- 116 shy

levels in many crops and locations in India [ICAR 1984] When 10-15 tonsha of farmyard manure were added to crops along with standard doses of chemical fertilizer the average yield for most crops increased However with rice wheat and maize there were many cases where yields did not change or else fell This may have been due to changes for the worse in farming practices but the results do indicate that the response to increased manure--and by implication to residue removal--are extremely variable The results from some of these tests are presented in Table 415

Table 415 Results of Long-Term Manuring Trials in India

Extra Grain Yield Using Manure (kgha) Crop Lowest Highest Average

Rice - 100 + 800 + 430

Wheat o + 600 + 290

Maize + 100 +1300 + 480

Millet o + 500 + 250

~ ICAR (1984)

These and related studies for India have shown that the financial cost to the farmer in lost crop production through burning animal wastes (and by analogy crop residues) is often less than the cost of using alternative fuels such as firewood [Aggarwal amp Singh 1984]

Prevention of Rain and Wind Erosion In the humid tropics rainstorms on bare sloping ground can remove very large amounts of soil Covering the ground with a layer of residue can reduce this loss by factors of 100-1000 For example trials in Nigeria established that on field slopes of 10 leaving 6 tonha of residue on the ground in periods when it would normally be ploughed bare would reduce annual soil loss from 232 tonha to only 02 tonha Water run-off was reduced by 94 because the residues both absorbed and retained the rainfall [Lal 1976] Where water is a limiting factor in plant growth residue mulches thus can increase crop yields by reducing moisture stress However the worst effects of water and wind erosion can be be mitigated without the need for residue mulches by terracing providing tree shelter belts and inter-planting and sequencing crops (and trees) so that the ground is nearly always covered by standing plants

- 117 shy

The economic costs of using residues instead of returning them to the land thus may be very high indeed or close to zero The costs depend critically on how much residue is removed and on the crop and farming system that is either practised now or could be practised if farming systems were to be adjusted to allow for greater volumes of residue removal Added to these issues are the various economic and opportunity costs of using residues as fuel rather than as animal feed or building material etc

Pellets and Briquettes

Densification of agricultural and forestry residues to briquettes or pellets is a method of expanding the use of these resources Densification increases the energy content per unit volume and thus reduces transport and handling costs The densities of residu~ briquettes are in the upper range for woods--namely 800-1100 kgm solid--wih a bulk density (ie for a sack or truck load) of around 600shy800 kgm Densification also produces a fuel with more uniform and predictable characteristics an important factor with medium to large scale energy conversion devices such as furnaces and boilers

For small-scale uses such as cooking the burning qualities of the fuel may be better than raw residues but this is not always so Some residue briquettes are smokey and hard to light or keep burning evenly--a factor which varies more with the briquetting process and briquette dimensions than with particular ligno-cellulosic residues Special designs of cooking stoves are sometimes needed to make the fuels acceptable Alternatively briquettes can be carbonized to produce a form of charcoal thus further reducing transport costs improving storage characteristics and providing a mOre easily adaptable cooking fuel

Since the processing costs are quite considerable densified residue fuels are normally intended for rural or urban industrial use and middle to higher income households in countries where either woodfuel prices are very high or residues are concentrated very close to demand centers Similarly since these residue fuels also show economies of scale densification is normally economic only at sites where raw residues are produced in substantial quantities eg centralized crop and food processing plant large cash crop estates saw mills logging centers and the like Supply estimates therefore are based simply on the volume flows through such plants

Densification Processes and Feedstock Characteristics

A variety of processing methods are available to make pellets or briquettes but they fall into two main categories low pressure systems such as manual or mechanical baling presses and high pressure systems which use rollers pistons or screw extrusion to produce relatively dense products Tandler and Kendis [1984] provide a thorough treatment of densification processes feedstocks and comparative costs

- 118 shy

The attributes of several densified residue feedstocks are summarized in Table 416 Table 417 presents the costs and other data on densification processes The most important characteristics for producing good quality pellets or briquettes are high lignin content low ash content and low to medium moisture content Lignin helps to bind the material together to make a durable product that will not crumble or powder during transport and handling If low lignin material is used higher pressures are needed to achieve binding Moisture contents below about 15 (wet basis) are essential to densification However more difficult residue feedstocks can be densified satisfactorily provided they are prepared and processed adequately For example more chopping or grinding may be needed before pressurization or higher pressures may be needed in order to plasticize small amounts of lignin into a binding agent Thus straw andrice husks which appear in Table 416 as poor feedstock materials can be densified satisfactorily with suitable processes

Table 416 Characteristics of Various Residue FeedstocKs for Densification

FeedstocKs Reason

Good

Poor

coffee hUSKS wood (not sawdust) bark cornstalks peanut she II s coconut shells bagasse (sugar cane)

straw rice husks cotton gin trash peat

high lignin high lignin low ash high lignin high lignin

high I ign in

low lignin high ash low lignin high ash low lignin high ash

Source Tandler and Mendis (1984]

Table 417 Characteristics of Denslflcatlon Processes and Products

Densificatlon Process

Energy Consumption of Equlpllent a

(KWht)

Product Density

(tem3)

Pel letlBr Iquette Production Rate

(tehour)

Range of Systell

Costs (US$OOOte h)

Cost per Unit Produced

(US$ OOOte h)

Product Characteristics

piston Extrusion Briquetting

30-60 NA NA

015-08 100 - 15

20shy 60 25 - 110

40 30

- 75 - 40

--

durable but breaks if over 25 mm long any length preferrably less than 25 mm long

Screw Extrusion Brlquettlng

50 - 180 NA 060 - 10 50 - 60 70 - 100 - feedstock moisture content may need to be low

Rol I Briquettlng 12 - 25 NA 10 - 45 75 - 170 40 - 75 - 25-50 mm size low denSity

45 - 90 170 - 300 30 - 40 - durable abi Ilty poor unless used binders

- p I I low-shaped

Pelletizing (Pellet Mill)

20 - 35 NA 20 - 60 130 - 300 30 - 60 ----

less than 30 mm high bulk denSity durable smooth easy storage handling conveying fuel

()

I

Cuber 15 - 30 NA 40 - 80 130 15 - 30 - lower density and durability than other extruder pellets

Bal ling 5 - 10 160 - 240 NA NA NA - less durable low density

Manual Presse NA NA 030 - 080 NA NA ---

vi Ilage-level production poor quality pellets binder is needed for durability

NA = not available ~ System energy requirements for the shredder dryer feeder and densifler generally range from 75 to 120 KWhte prOduct

~ Tandler and Mendis 119841

- 120 shy

Energy Content and Costs

Table 418 provides heating values and some indicative costs for the major residue briquettes based on studies in Ethiopia [Newcombe 1985] At typical moisture contents of 10 most briquettes contain 16-18 MJkg net heating value (175 MJkg on average) or some 10-20 more than firewood at its typical air-dried moisture content This compares to an average 14 MJkg for the same residues in non-briquet ted forms

Table 418 Average Net Heating Values and Costs of Briquetted Residues

Net Heating Cost of Value al Delivered Energy

Feedstock (MJkg) (USfIMJ)

Coffee Res i due 176 MJkg 042

Bagasse 173 MJkg 052

Cotton Residue 178 MJkg 052

Cereal Straw 171 MJkg 053

Sawdust 177 MJkg 055

Cereal Stover 187 MJkg 068

al Net heating values assume 10 mcwb

Source UNDPlWorld Bank (1984b)

Briquettepellet costs will vary considerably according to the densification process the scale of processing and the original biomass feedstock Collection costs for harvesting feedstocks such as cotton stalks and cereal straws may be considerable but with residues that arise as by-products in crop processing plants (eg coffee bean husks) the feedstock costs are negligible unless there is an opportunity cost for alternative uses

Table 419 gives some costs for harvesting densifying storing and packing various residues in Ethiopia [Newcombe 1985J The economic costs range from US$25-32ton unbagged at the processing plant and U5$26-34 per GJ energy content bagged and delivered 300 km to the market These costs are low compared to fossil fuel alternatives The ready to burn costs at the market are equivalent to unprocessed crude oil (58 GJbarrel) of only US$15-20 per barrel Transport and bagging

- 121 shy

in the Ethiopian case studies make up 38-44 of the economic cost delivered to the market

Table 419 Production Cost Estimates for Commercial Scale Crop Residue Briquetting in Ethiopia

(USS (1983)ton of product)

Residue (1) (2) (3)

Corn amp Wheat amp Cotton Sorgurn Barley

Stage of Production Stalks Stover Straw

Harvesting Capital charges Energy amp lube Maintenance ampother Labor

Grinding

Brlquetting Capital charges Energy amp lube Maintenance ampother Labor

Storage etc Financial cost ex-plant Economic cost ex-plant Economic costs of transport and bagging etc

Bagging (40 kg sacks) Transport I Handling at each end

Economic cost delivered to market

Net heating value MJkg Moisture content ~ (wb)

Economic cost per energy unit del ivered to market USSGJ

723 (422) (135 ) (150) (016)

1180 (556) (1 76) (437) (011)

10 2005 2502

1941

(338) (1403) (201)

4443

173 ( 12)

2257

1903 (1040) (411) (432) (020)

144

854 (237) (52S) (080) (012)

088 2989 3215

1941

(338) (1403) (201)

5156

150 (15)

344

1085 (239) ( 1 64) (640) (042)

144

8S4 (237) (S2S) (080) (012)

088 2171 2735

1941

(338) ( 1403) (201)

4676

174 (15)

269

a Transport 22 ton trucks over 300 km of deteriorated paved roads to Addis Ababa

Source Newcombe [19851

- 122 shy

G ANIMAL WASTES

Direct Combustion

Animal wastes are either burned directly as dried fuel or processed in a digestor to produce biogas and a fertilizer slurry Like crop residues animal wastes are vital fuel resources in many wood-scarce areas of developing countries for rural and urban low-income groups In India an estimated 12 million tons of cattle dung were burned as fuel in 1918-19 [Natarajan 1985]

Since a mature bovine produces roughly 5-1 tons of fresh dung annually with an oven dry weight of 13-11 tons and an energy content of 16-22 GJ (or up to half a ton oil equivalent) the potential fuel supply can be large wherever animals are kept for draft power as well as meat milk and hides etc But the availability of this material as fuel is a much more pertinent factor Apart from questions of whether animal wastes should be removed from the land dung availability will be high only when (1) animals are stalled or corralled for substantial periods of time or (2) when people are prepared to spend time collecting it from the fields and pastures etc Only the poor women who collect dung for sale and the servants of the rich are normally prepared to do the latter In village level studies it is also of vital importance to allow for the distribution of animal ownership by household and customs of dung barter and collection rights on common land etc since these factors have a profound bearing on who can and cannot burn dung as a fuel (or benefit from its conversion in a biogas plant) Supplies may also vary greatly by season since dung cannot be collected from the fields during prolonged wet weather

Table 420 presents some data on annual dung production wet and dry for a range of average animals as well as the nitrogen content of animal dung These values could be used for rough order of magnitude estimates but always should be checked against local data The need to use local information is underscored by the enormous range of production figures that has been found in detailed Indian surveys which attempt to establish the availability and costs of dung for the countrys biogas program For example although the all-India mean figure for wet dung production by cattle is 113 kgday (41 tonyd the mean figure for different states ranges from 36 kgday (Kerala) to 186 kgday (Punjab) [Neelakantan 1915]

Table 420 Manure Production on a Fresh and Dry Basis for Animals In Developing Countries

Fresh Manure Basis Drl Manure B8Sls

Animal

Fresh Manure per 1000 kg lIveweight

(kgyr)

Assumed Average Liveweight

(kg)

Fresh Manure Production Assumed per Head (kgyr)

Assumed Molsshyture Content of Fresh Manure (percent)

Dry Manure Production per Head (kgheadyr)

Nitrogen Content Percentage of Drl Matter

Solid and Sol id Liquid Wastes Wastes Only

Cattle 27000 200 5400 80 1000 24 12

Horses mules donkeys 18000 150 2700 80 750 17 1 bull I

Pigs 30000 50 1500 80 300 315 18

Sheep and goats 13000 40 500 10 150 41 20 ~ N W

Poultry 9000 15 13 60 5 63 63

Human feces without urine 40 to 80 50 to 100 66 to 80 5 to 1

Human urine 40 to 80 to 25 kg 15 to 19 dry so I I dsyr (urine only)

Sources Bene et al [19181 and Hughart [19191

- 124 shy

The heating value of dung is usually lower than crop residues because it contains more inorganic material Fresh dung is often contaminated with earth or grit while it is often mixed with straw and other residues when it is dried and patted into dungcakes One set of detailed measurements from Thailand put the gross heating value of fresh dung oven dry basis at 118 MJkg for buffaloes 128 MJkg for cows and 149 MJkg for pigs [Arnold amp deLucia 1982] When air-dried to 15 moisture content (wet basis) the respective net heating values are 86 MJkg 94 MJkg and 112 MJkg using the formula for firewood presented in Chapter I Other estimates in the literature range from 10-17 MJkg although it usually is not clear whether these refer to air dried or oven dry material

Biogas

The biodigestion of dung and residues to gas appears to offer an enormous potential for bringing cooking heat light and electric power to the villages of the Third World Yet it is discussed here only briefly for three reasons First the technology is peculiarly dependent on many specific local circumstances which favor or work against its success and therefore can be assessed only by site-specific studies Second there is a vast literature on the topic which can assist in such studies especially in India China Thailand and a few other countries which have pioneered the biogas digestor (see for example the recent major study by Stuckey [1983]) Third due to very high failure rates--among small family size digestors--it is not yet a technology that appears suitable for household energy use The main successes have been with village-scale plants that run irrigation pumps and other machinery as well as provide household fuel and large-scale digestors attached to agro- and food-processing plant and animal feedlots

There are serveral key points to note about the technology as it applies to household use

3a Small family-size systems of 3-4 m capacity have experienced extremely high failure rates Of the 300000 units installed in India almost half are routinely out of order [FAO 1985b] A 1978 survey in Thailand found that 60 of the family-size installations were non-operational [UNDPWor1dBank 1985b] and experience has been equally discouraging in other ASEAN countries One of the main reasons for these high failure and abandonment rates is that biogas digestors are labor intensive and require a high level of management and experience to operate successfully

b Costs are either high for materials as in the Indian-style steel drum systems or in skilled labor as in the buried masonry systems pioneered in China Recent data for Indian systems give investment costs of US$230 and US$335 ($1981) for 2 m3 and 4 m3 family-size units respectively while dung from

- 125 shy

2-3 and 4-6 animals is needed to keep them operating Families who could afford these investments and own as many cattle are often in the income group which is shifting towards fossil fuels for convenience or the sake of modernity They are likely to invest in biogas only if there are clear advantages outside the area of household energy such as using the gas for power generation andor irrigation pumping

c Perhaps more than for any other topic discussed in this handbook there 1S a dearth of reliable and comparable information on biogas systems except in a few specific locations from which generalizations cannot be made This point has been noted in many studies including the UNDPWor1d Bank assessment by Stuckey [1983] cited above The Stuckey assessment calls for a comprehensive and systematic global biogas program to provide reliable technical economic and social data to use in unravelling the uncertainties surrounding biogas use in developing countries

- 126 shy

CHAPTER V

ASSESSMENT METHODS AND CASE STUDIES

Project analysts and planners concerned with household energy need to identify the key issues and options for the sector as a first step in identifying policy and project goals To do so they must draw on a wide variety of information not only about patterns of energy resources supplies and demand but also wherever biofuels are important about related areas such as agriculture forestry the commercial wood trade transport costs and manufacturing capabilities The socioshyeconomic conditions and attitudes of families are also critical components of many types of energy assessments However the main requirement is to keep a clear eye on the main principles which can so easily be overlooked in the welter of details

This chapter presents some broad methods of analysis and the principles that underlie them The emphasis is on biofuels since these raise questions which may be unfamiliar to many readers The emphasis is also on first-order appraisals from available information which aim to identify the main issues and opportunities for change through policies projects or other types of intervention Preliminary appraisal methods must be employed in all analyses and so are worth discussing here The chapter does not consider in any depth the great variety of other assessment methods and analytical approaches that are required to turn preliminary scoping studies into well formulated policies and projects The focus therefore is on ways to identify major policy and technical issues and select options for further study rather than detailed project assessment

With this aim in mind the chapter begins with a brief review of data sources The limitations of the information available about energy resources and supply and demand for the household sector have a great bearing on the types of methods that can be used The simplest and most aggregate approaches to projecting biofuel resources supplies and demand therefore are presented as a means of identifying policy priorities These approaches are then refined in order to provide greater reliability and value

B DATA SOURCES

Demand Data and Data Sources

As we saw in Chapter II there are four main sources of household energy data on the demand side

- 127 shy

a National Energy Balances Usually developed annually although household data is limited highly aggregated and often unreliable for biofue1s Regional differences such as in fuel abundance or scarcity are rarely noted

b National Household Expenditure Surveys Usually large nationally representative surveys with a reasonable degree of disaggregation such as for type of fuel used and main categories of household including income household size rural-urban location and sometimes region Data are often based on recollection and so may be unreliable and are given in terms of cash expenditure rather than physical quantities (although the latter can usually be obtained from the survey source) bull

c National Household Energy Surveys Where they exist these are usually by far the richest source of disaggregated data As well as breakdowns provided in (b) they may also give data on attitudes preferences and technologies used

d Local Micro Surveys These can provide excellent data on energy use and supplies as well as the diversity of demandsupply patterns attitudes and behavior They may also provide information on the total system of biomass resources flows and consumption (agriculture livestock etc) critical inputs to the system and differences in these respects between various socio-economic classes Extrapolation to the regional or national level is rarely valid and should be avoided unless there is evidence that the survey locations are typical or there is no other information to go on

methods Table 51 provides a

and associated problems checklist of data needs assessment

in the analysis of cooking energy the major end-use in the household sector It draws on the material presented in previous chapters

In assembling this information at any level of aggregation some cardinal rules are worth bearing in mind These also apply to supply data which is discussed in the next section

Do not be be guided by averages it is often the variation and the extremes that matter most since they can (1) point to the locations where fuel problems are greatest or likely to become so and (2) give clues to how people have adapted to different conditions (eg burning more crop residues or purchasing nonshytraditional fuels where woodfuel resources are particularly scarce)

- 128 shy

Table 51 Cooking Energy Demand Analysis Oata Needs Methods and Problems

Data Methods Problems

Household amp Numbers in National population categor i es used statistics demographics below surveys

Fuel use Per capita amp Surveys Measured rather than recall data Uncertain heat per househo I d values for biofuels (moisture content etc)

By household Surveys Variation by household category culture and category (rural diet firestove management technologies used urban Income household size etc) By fuel Surveys Multiple fuels ampequipment multiple uses of

cooking heat (especially space heating) Technologies Efficiency by Testing ampsurveys Uncertain estimates often better to compare ampefficiencies equipment type specific fuel use for technologies (existing amp hence improved)

Equipment Expense ownership surveys

Useful heat for UH =fuel use Technology changes may not give estimated fuel cooking 2 x eff Iclency savings due to changes in management multiple

relative fuel uses etc use (RFU) for RFU observed technologies directly

Technologies see I Prob I ems I Observation Fueltechnology preferences ampaversions often ampcultural anecdotes for non-energy reasons (smoke safety Insect factors control convenience etc) Technologies Capital amp repair Relative costs First cost may be major barrier even if ampcosts costs Lifetime of utilized heat low life-cycle costs Varying time

Fuel prices -= pr i ceeff I cshy horizons for Investments Cost uncertainties Efficiencies or ency or price eg mass production v test models RFU x RFU li feshy

cycle costs

Do not i because it has not been measured (or you cannot measure it qualitative information is often as important as quantitative data in forming assumptions

Your data requirements must be driven by your problem which often means that you need less data than you think

Distrust the simple single answer as there is usually a range of interrelated solutions some of which may lie outside the energy sector

Make your assumptions explicit so that you or others can change them as the data or ideas improve

- 129 shy

Rural inhabitants are the best judges of what is good for them especially where biomass resource and consumption systems are fairly complex

In many situations and types of assessment the single most important rule to bear in mind is that existing demand patterns will change with time They will be adapted through feedback to changes in supply and resources This is well recognized for modern fuels where income prices fuel availability etc are known to be key variables which affect the level and choice of fuels used Many assessments of traditional fuels on the other hand assume that existing patterns of demand are immutable and will persist through every reduction in available resources

In most cases though there will be no information on which to judge the type or scale of these adaptations The lack of adequate time series data on household energy parameters (and their relation to other factors) means that one must work without any clear sense of history of past experience and must instead include the concept of future change as an assumption (or variety of assumptions) This has important implications for all that follows It means that assessments must usually be based on what if scenarios or projections which may also be normative in character That is projections are made from starting data (or assumptions) about the present by making further assumptions about natural rates of change (eg in response to rising fuel prices or firewood scarcity) or certain deliberate policy andor technical changes (eg the introduction of so many improved stoves each year) Projections of this kind are particularly valuable for policy formulation and project selection since they show in a transparent way the likely (estimated) outcome of policy actions Some illustrations are given below

Supply Data

Information about household biofuel supplies normally must be estimated from consumption data as described above Actual or potential supply volumes are very rarely recorded by household consumption surveys The same is true of modern fuels such as kerosene and LPG except for the most aggregate or total data As discussed in Chapter III electricity and piped gas are the only energy sources for which data on the household sector is dissagregated by region or type of household

Equally important are data on biofuel resources potential supplies and available or economic supplies allowing for competing uses There are two main kinds of resource information to consider-shyinformation on tree resources and information on residue resources

a Tree resources These include all types of tree formations such as forests and woodlands single tree resources (ie trees dispersed through urban and agricultural ecosystems) and managed forests (ie plantations and woodlots etc) The

- 130 shy

important quantities that may be required for an assessment are (1) 1and areas under forests and plantations (2) the standing

3stock (m hal (3) the gross sustainable yield or Mean Annual Increment (m hayr) and (4) the fraction of both (2) and (3) that is or could be available as woodfue1 for a given market allowing for physical accessibility competing uses such as timber and poles environmental considerations and the costs of preparing and transporting woodfuels This type of data usually is required for major regions within a country and with breakdowns by land type

Many developing countries now have data on land use and land types which include estimates of the standing stocks and annual yields of trees and other woody plants Some typical stock and yield data were presented in Chapter IV This type of information is normally held by the government forestry surveyor planning departments (or appropriate academic units) and is collected by a combination of satellite imagery aerial survey and ground observation Data on woodland stocks and yields for most developing countries are also published in the regional volumes of the Tropical Forest Resource Assessment Project conducted by the UN Food and Agriculture Organization (FAO Rome) and the UN Environment Program (UNEP Nairobi) Although estimates are approximate in many countries the quality and quantity of data are steadily improving as recognition of their importance to biofue1 planning increases

b Residue resources These include woodfue1s crop residues and animal wastes which are generally flow resources rather than the stock plus flow resources discussed above For woodfue1s the major resources are concentrated and include logging and sawmill wastes Data may be difficult to obtain unless there has been a recent survey of commercial forestry and timber operations For crop residues and animal wastes the main sources of data are agricultural statistics or occasional agricultural and animal censuses Data from these sources on crop areas their location and crop yields can be combined with the residue yield factors given in Chapter IV to estimate total residue production A similar approach can be used for animal wastes using data on the number and size of domestic animals and daily dung production (see Chapter IV) Wherever possible local data should be used since there are considerable local variations in crop yield and cropresidue ratios Estimating the amount of this material that is or could be available as an energy source allowing for alternative uses is much more difficult Local micro surveys or specific studies on this point may provide some guidance

Table 52 provides a checklist of data needs assessment methods and associated problems in assessing biofue1 resources and supplies

- 131 shy

Table 52 Woodfuel Resources and Supplies Data Needs Methods and Problems

Data Methods Prob Iems

land use

Wood resource stocks ampyields (closed ampopen natural forest bushscrubland single tree managed forests ampwoodlots)

Physical amp economic accessibility

Resource ava II abll Ity (allowing for competing uses)

Costs prices ampeconaics (firewood)

Costs prices ampeconaics (charcoal)

Area of main land types by region

Stndl~g stock (m II ha) amp sustaina~le yeld (m yr III hayr) by resource type

Fraction of stock currently accessed reasons for I I mI ted access

Accessibility under different conditions (population density cost etc)

Volumes for tllllber poles etc Fraction of resource now used for woodfuels Actual woodfuel take

ConIIIerc i a I harvest costs producer prices transport amp marketing costs ampprofits Non-commercial local practices ampattitudes

As above plus costs amp efficiencies of ki Ins

National International statistics

As above

Gross stock amp yields x accessibility = net stock amp yields

Physical amp economic analYSis

Forestry amp commercial statistics local surveys

Deduct compet I ng uses multiply net stockyield x fraction avai lable Use actual take

Estimate market and economic costs aval I able resources at these costs Repeat for future costs amp prices

As above

Data quality varies widely by country

As above large variation by type (eg age of woodlands species) soilcllatlc region management practices

Uncertain data large local variations Most data Is for commercial timber

As above Future estimates especially uncertain use sensitivity analysis

As above

Uncertain data Much fuelwood (amp charcoal) Is produced amp marketed by the informal economy

Poor data for noncommercial coilection variable responses to abundancescarcity

As above

- 132 shy

C SIMPLE SUPPLY-DEMAND PROJECTIONS

Forecasts of energy demand and supply are well recognized as a valuable tool for identifying imminent problems in the sector In this section we review the value methods and precautions that must be considered in making the simplest first order projections of woodfue1 demand and supply

Constant-Trend Based Projections

A useful initial analysis for the biofue1 sector is to assume that there are no feedback mechanisms at work so that there is no change in unit consumption and demand grows in line with population growth One also assumes that nothing is done to increase available supplies and resources through efforts such as afforestation Projections can be made at any level of aggregation at the national or regional levels or for a particular town or village

The main uses of such projections are (1) to identify any resource problems and (2) to ascertain if a problem does exist the degree of future adaptation required to bring supply and demand into a sustainable balance If there is a problem the projection is merely a starting point for further work since it describes a future that is most unlikely to come about in practice

Table 53 presents a sample projection The basic data on consumption population and resources are given below the table and are used in subsequent projections in which the methodology is refined The calculation method is also presented with the table Essentially consumption grows with the population at 3 a year and supplies are obtained from the annual wood growth and clear felling of an initially fixed stock (area) of trees We assume at this stage that there is no use of agricultural residues or animal wastes as fuels

The starting conditions for the projection reflect the situation in many areas of the developing world wood consumption exceeds wood growth so that supplies are partly met by cutting down the forest stock In the first few years the rate of resource reduction is small (only 18 annually for the first forecast period) It may not be noticeable to local residents or may appear less threatening than other problems of survival Unless adaptations which slow or halt the decline have large perceived benefits andor low costs they are unlikely to attract much interest However since demand is assumed to rise exponentially the resource stock declines at an accelerating pace and eventually falls to zero (in this case by the year 2007)

- 133 shy

Table 53 Constant Trend-Based Projection Wood Balance

1980 1985 1990 1995 2000 2005

Standing stock 000 3 17 500 16010 13837 10827 6794 1520

Fuelwood yield 000 3yr 350 320 278 217 136 30

Consumption 000 3yr 600 696 806 935 1084 1256

Deficit 000 3yr 250 376 529 718 948 1226

(Population ooos) (1000) ( 1 159) (1344) (1558) (1806) (2094)

Assumptions

Fuelwood yield 2 of standing stock (Standing stock 20 m3ha) Population 1 million in 1980 growth at 3 per year Consumption 06 m3caPltayear Deficit is met by felling the standing stock

Calculation method

Calculations are performed for each year (t t+l etc) taking the stock at the start of the year and consumption and yield during the year

Consumption (t) =Reduction in stock (t t+l) + Yield in year (t)

Stock (t) - Stock (t+1) + M2 x [Stock (t) + Stock (t+l)]

where M = YieldStock expressed as a fraction (002 in this case)

Hence to calculate the stock In each year

Stock (t+l) x [1 - Ml2] = Stock (t) x [I + M21 - Consumption (t)

Such a picture of the long term is unrealistic at best As wood resources decline ever more rapidly wood prices and collection times would rise and consumption would be reduced by fuel economies and substitutions of other fuels

Projections with Adjusted Demand

A useful next step is to examine reductions in per capita demand to see how large they must be to reduce or halt the decline in wood resources The adjustments can then be related to policy and

- 134 shy

project targets such as improved stove programs and substitutions of other biomass fuels or petroleum-based cooking fuels for woodfue1s

An exercise of this kind is shown in Table 54 using the same basic assumptions used in Table 53 The calculation method is quite simple The population (A) is divided into categories of fuel and equipment users in this case for cooking (B) Estimates are made of the specific energy consumption of each category (C) Total energy for each category (0) is the product of (A) x (8)100 x (C) Finally total wood energy is converted to a wood volume (E) Apart from demographic information the only data required for the projection are those shown in the first column of (A) (8) and (C) plus rough information on fuel savings that can be achieved by economies and more energy efficient equipment

In this example three main kinds of wood saving are considered

a Substitution of improved stoves for open fires (8) This may result from market forces increasing urbanization and incomes or a proposed program for introducing improved stoves The rate of substitution assumes a logistic curve for the proportion of wood users employing stoves (F) From these assumptions the rate of stove introductions can easily be calculated (F) The implied stove program expands fairly steadily to 1995 and then slackens off as saturation in stove ownership is approached Alternatively annual targets for stove introductions can be used to derive the data in (B)

b Substitution of wood by crop residues (in rural areas) and petroleum products (in towns) at a gradually accelerating pace The former change is a common response to wood scarcity the latter to urbanization and rising incomes Substitution into petroleum cooking fuels (and electric cooking) may also be the result of policy choices for urban areas facing woodfuel deficits as occurs in some developing countries today

c Reductions in specific fuel consumption by all user categories The largest reductions (40 over the 25-year period) apply to open fires since the scope for economies is greatest here For the stove and residue groups the equivalent reductions are 30 and for the petroleum product group 17 In all cases much of the reduction could be due to the use of more efficient cooking equipment such as aluminum pots and pressure cookers (see Chapter III) Some reductions could also be due to progressive improvements in stove efficiency and the introduction of stoves for use with crop residues perhaps through pelleting and briquetting

- 135 shy

Table 54 Basic Projection Adjusted for Demand

1980 1985 1990 1995 2000 2005

(A) Population (ooos) 1000 1159 1344 1558 1806 2094

(B) Fuel ampeguipment use (percent) Wood 80 78 72 66 56 45

open fire 75 663 504 33 196 10 stove 5 117 216 33 364 35

Residues 10 11 14 17 22 25 Petroleum products 10 11 14 17 22 30

(C) Per capita consumption (GJ) Wood 90 86 76 62 50 37

open hearth fire 93 93 90 83 73 56 stove 46 46 44 41 37 32

(0) Total consumption (000 GJlr) Wood 7205 7770 7373 6375 5016 3518

TOTAL 8505 9389 9669 9421 9267 8753 Totalcapita GJyr 851 810 719 605 513 418

(E) Wood consumption 000 m3yr 600 647 614 531 418 293

(F) Supplementarl data Wood users with stoves (J) 63 15 30 50 65 78 Increase in stoves over preshyceeding 5 years ooosyr 34 62 90 57 30

For calculation method see text

Assumptions As for Table 53 plus Fuelwood of 600 kgm3i 20 MJkg (both oven-dry basis) Stove introduction rate assumes 5 persons per household

- 136 shy

These adjustments cut annual wood use in half over the projection period The effect of this change on wood resources is shown in Table 55 The reduction in stock over 1980-2005 is now only 37 and equally important consumption and resources come close to being in balance by the end of the period The catastrophe of total deforestation has been averted

Table 55 Basic Projection Adjusted for Demand Wood Balance

1980 1985 1990 1995 2000 2005

Standing stock ltogo m3) 17500 16103 14479 12960 11777 11082 Wood yield lt000 m ~yr) 350 322 290 259 236 222 Consumption (o~ m Iyr) 600 647 614 531 418 293 Deficit lt000 m Iyr) 250 325 324 272 182 71

Assumptions As in Table 53 consumption from Table 54

The projection presented in Table 55 may also be considered unrealistic since wood savings continue to accelerate at a time when demand and resources are brought into balance However this objection misses the point of projections of this kind They are not intended to forecast one particular future as much as to explore alternative futures and the role of policy interventions in achieving these alternatives Thus their purpose is to explore the effects of given changes--to ask what if--and hence to help select the policies and projects which aim to bring about those changes The realism of a scenario lies in the likely timing scale and successful adoption of the interventions recommended and can only be judged after the fact For this reason it is always valuable to make a variety of projections to illustrate the implications of different policy initiatives and outcomes

Projections with Increased Supplies

Woodfuel deficits may also be reduced by a variety of measures which increase the supply of woodfuels or alternative biofuels Woodfuel supplies can be increased by more productively managing existing forests planting trees in rural areas for fuel or multiple purposes or setting up periurban plantations For example logging and sawmill wastes may be utilized economically Many agricultural changes can be made to augment supplies of crop residues or animal wastes so that they can be used more extensively as fuels without competing with other essential uses The briquetting and pelletizing of agricultural residues often can make these fuels more widely available at economic prices

Targets for these additional supply options can easily be set by estimating the gap between projected woodfuel demand and supplies since the objective is to eliminate woodfuel deficits Various mixes of

- 137 shy

supply options can be considered with different levels of demand reduction so that together they achieve a balanced projection Examples of balances with a variety of additional supply outputs are presented in the case studies of Section E

Projections Including Agricultural Land

A major shortcoming of the projections discussed above is that they ignore the effects of the expansion of agricultural land In most developing countries the spread of arable and grazing land together with commercial logging in some places has been a much mare important cause of tree loss than the demand for woodfue1s (see Chapter IV)

The effects of agricultural land expansion are illustrated in Table 56 using the same hypothetical system as before Assuming no increase in agricultural productivity farm land increases by 3 annually or the same as the growth of population This expansion is alone responsible for a 63 decline in woodland area and wood stocks over the period of analysis If much of the land is cleared by felling and burning--a common practice in many areas--this wood would not contribute towards meeting some of the demand causing additional pressures on the forest stock and leading to their very rapid decline On the other hand if one assumes that all the wood from these clearances is used as fuel-shyas in Table 56--then the wood made available from land clearance and natural regeneration would be sufficient to meet a 2 annual growth in fue1wood demand without resorting to tree cutting for fuel in the remaining woodland areas

This simple example underlines the critical importance of including agricultural parameters in wood resource and demand projections and the need to establish whether trees and woodlands that are cleared for farming are burned in situ or are used as fuel and timber - -shy

Projections Including Farm Trees

A particularly important source of supply often ignored ln these types of projections is the fuelwood from trees growing on farm lands to produce fruit forage small timber shelter shade or fuelwood itself These represent a major source of fuel for many rural inhabitants and provide another very important reason for including the agricultural system in projection models

An example of the potential contribution of farm trees to fuelwood supply is provided by a number of FAOUNDP Tropical Forest Resource Assessments for East Africa In addition to timber and construction poles these assessme3ts revealed that farm trees can provide on average as much as 05 m of fuelwood a year per hectare of total farmland in some regions (see Table 57) [Kamweti 1984] This is more than the gross yields from the woodland uses in the projections above

- 138 shy

Table 56 Projection Based on Expansion of Agricultural land

1980 1985 1990 1995 2000 2005

(A) Areas and stock Woodland area (000 hal 875 795 703 596 472 328 Agrlc area (000 hal Standing stock lt000 m 3)

500 17500

580 15907

672 14061

779 11920

903 9439

1047 6562

(B) Wood avai labl itl (000 mLr)

New agricultural land 300 348 403 467 542 628 Woodland yield 347 315 277 234 183 125

TOTAL 647 663 680 701 725 753

(C) Consumption and WOOd Balance (000 mLr)

Consumption growth 2 pa Consumption 600 631 663 697 732 769 SurplllsOeflclt (+-) + 47 + 32 + 17 + 4 - 7 -16

Assumptions Agricultural area 05 hacapita Population as in Tables 53 - 55 Consumption growth as shown All wood from land cleared for agriculture is used as fuel Wood availability equals stock from land clearance plus yield of remaining woodlands ie no trees are cut for the direct purpose of providing fuel

Furthermore farm trees are fully accessible to the local consumers of their products The accessibility of forest and woodland resources is rarely 100 and is usually much less than this because of physical reasons (remoteness from consumers difficult terrain) economic reasons (transport costs to major demand centers) or legal reasons (prohibitions on access to or cutting within game and forest reserve) Consequently available or net yields of fuelwood are normally much less than the gross yields used in the examples above The present accessibility of these resources and likely changes in population density and location costs and prices and infrastructural factors such as road building are often critical factors to consider in making projections of the kind discussed here However these factors are difficult to quantify as they are subject to great uncertainty

- 139 shy

FIGURE 51 Indices of Forest Stocks Varying On-farm Fuelwood Production and the Rate of Decline in Per capita Fuelwood Consumption

Annual Reduction In Per Capita100r-

Wood Consumption

~~5~ 43 2

1 On-farm Wood 01 m3hayr

Annual Increase 0

0 O~________L-________~________~________-L________~

1980 1985 1990 1995 2000 2005

100r--__bullbull

~~====3--- 2

1

0

On-farm Wood 04 m3hayr Annual Increase 2

o~--------~--------~----------~--------~--------~ 1980 1985 1990 1995 2000 2005

Common Assumptions Annual Population Growth 3 Annual Increase in Agricultural Productivity 3 (Ie Constant Agricultural Land Area)

World Bank-307364

- 140 shy

The effect of including on-farm fuel wood production in the wood balance of our model system is shown for two cases in Figure 51 In both cases agricultural productivity grows in line with population so that the area of agricultural land remains constant In the top figure on-farm wood production is initially low and per hectare yields do not increase Consequently if the decline of the forest stock is to be arrested per capita fue1wood demand must fall by about 5 annually In the lower figure on-farm production is initially quite high while average per hectare yields grow at 2 annually reflecting a fairly vigorous programme of rural tree planting Now the forest stock is stabilized at close to its initial level with only a 3 annual decline in per capita fue1wood consumption

All the examples in this section illustrate the necessity of elaborating on even the simplest wood balance projections Without the progressive addition of the concepts outlined above the projections will be of little value and may actually misdirect the process of selecting and examining policy options

D DISAGGREGATED ANALYSES

In practice the models and projection methods used for national planning cannot be as aggregated as in the examples presented above The diversity of the basic projection parameters and their trends makes it necessary to use some degree of disaggregation both for demand and supply projections

Aggregated models also are limited in that they can be used only on a limited number of well-defined target subsystems or regions within the country The target may be a major urban demand center a rural area experiencing rapid population growth or inward migration an area of rapid agricultural expansion or a region that is suitable for afforestation or rural tree-planting schemes The target may be as small as a single village

Demand Disaggregation

As discussed in Chapter III household energy demand and the mix of fuels employed vary greatly by settlement size household income availability prices and other factors Different household groups also vary in the opportun1t1es constraints and costs they perceive are involved in changing their energy use and supply patterns Therefore national demandsupply projections and balances wherever possible should be derived from disaggregated projections for the major types of households The level of disaggregation of these projections must be a judgement for the analyst based on available data and the degree of difference existing between the sub-groups

- 141 shy

Another major criterion in determining the optimal level of disaggregation is the computational effort involved For the examples presented above results were obtained quite rapidly by using either a programmable calculator or simple computer programs For disaggregated models computer spreadsheets or software designed specifically for analyses of this kind are almost a necessity A good example of the special software which has been installed in a number of developing countries is the LEAP (LDC Energy Alternative Planning) system developed by the Beijer Institute Stockholm and the Energy Systems Research Group Boston Massachusetts USA On the demand side LEAP provides for extensive disaggregation by energy consumption groups ownership of energy equipment specific fuel consumption and efficiencies On the supply side LEAP has sophisticated modules for the modern energy sector land use and land types and the resource and production characteristics of a large range of biofuels

Resource and Supply Disaggregation

The need to disaggregate biofuel resources and supplies is illustrated in Table 57 which shows population land use and types and fuelwood production characteristics averaged for six East African countries (Ethiopia Kenya Malawi Somalia Tanzania and Zambia) Gross fuelwood yields vary by a factor of 17 from the least to the most productive regions and land types Furthermore while the average yield per hectare ranges from about 50 to 600 kgyr the average yield per capita is not related to this quantity because of the large variations in population density compare for example Zones 1 and 6

The main lesson to be learned from the type of regional breakdown presented in Table 57 is that woodfuel deficits as well as demand and resources usually vary considerably This variation is often the result of differences in population density and agricultural land area which are themselves related to the basic biological productivity of ecosystems Thus in Table 57 one sees that on average sustainable woodfuel yields probably exceed deman~ in all but two areas the dry savanna (Zone 3 with a yield of 073 m hayr) and the heavily populated highlands (Zone 6 with a yield of 039 m3hayr) These are clearly the areas most likely to be suffering severe deficits and woodland depletion and hence are priority areas for more detailed assessments or project development However other areas may well be in a similar plight since the table shows only the gross yields and not the net yields allowing for accessibility Note also that there are large differences between the zones in the proportion and growth rates of agricultural land and hence in on-farm wood supplies

- 142 shy

Table 57 Population and Fuelwood Data by Land Type Averages for East Africa 1980

Land type 2 3 4 5 6

Population 42 84 374 77 21 402 Total land area 265 98 367 120 71 79

Population density 30 160 192 122 56 964 (personskm2)

Area of land by type ( total area)

Closed forest 02 36 15 31 126 51 Woodlands 18 40 37 96 121 28 Bushlands 88 306 219 322 277 177

Scrublands 464 543 296 121 60 222 TOTAL 572 925 567 570 584 478 (Agriculture) (42) (64) (167) ( 140) (81) (336)

Gross fuelwood yield ie without deductions for accessibility (m3hayr)

Closed forest 10 20 10 15 18 25 Woodlands 04 06 08 10 12 12 Bushlands 015 04 03 075 08 085 Scrublands 005 015 01 025 03 03 (Farm lands) (02) (035) (025) (04) (045) (05) (PI antations) (20) (100) (50) (140) (150) (160)

Note standing stock = 80 x gross yield

Average yield per total area m3hayr 0046 0300 0141 0414 0613 0379

Average yield per capita m3yr 150 188 073 340 110 039

Land type

1 Desertsub-desert 2 Warm humid lowlands 3 Dry savanna

4 Rapid agricultural expansion 5 low populationslow or no

agricultural expansion 6 Heavily populated highlands

Source Kamweti [19841

Altitude (m)

200-1000 0- 500

500-1500 1000-2000

1000-2500 1500-3000

Rainfall (mm)

lt400 500-1000 500- 900

800-1200

1 000-1 300 lt1200

- 143 shy

51 It is clearly beyond the scope of this handbook to design micro-computer spreadsheet data bases and models to encompass regional disaggregation and its complications However this process would call for no more than simple arithmetic and algebra and an ordered approach The basic formulae for making projections are presented in this handbook or can be derived by common sense Alternatively packaged systems such as LEAP can be used

E CASE STUDIES

52 To summarize the methods and concepts outlined above this section provides a case study of a target analysis for household energy demand and supply The example is based on an analysis of supply options for the household sector of the Antananarivo district (Faritanytt) of Madagascar [UNDPThe Wor1d Bailk1985a]

53 Per capita and total fuel consumption were estimated by surveys of a few main regions of the country Demographic data also were assembled The results of this demand analysis for woodfue1s are summarized in Table 58 although data on modern fuels also were collected Note the large consumption differences between the regions and the fact that the energy unit is tonnes woodfue1 equivalent rather than GJ etc Although this may upset energy analysts it is a descriptive term useful for politicians and economic planners in countries where woodfue1s dominate the energy market It is also more easily understood and utilized by foresters and transport planners

Table 58 Household Woodfuel Use in Urban and Rural Centers of Madagascar

(A) Per capita woodfuel consumption (kgwoOd- eq iva lent per year)

Highlands bewlands Overall Fuel Urban RUfl81 Urban Rural

Firewood 70 550 100 365 Charcoal 140 0 70 0

(B) Total Woodfuel Consumption (thousand tonnes wood equivalent)

Highlands Lowlands Overall Total

Average Both fuels

548

Firewood Charcoal Total

2344 1148 3491

1482 362

1844

3826 1510 5336

Source FAOCP Fuelwood Project Preparation Mission (1983) and UNDPWorld Bank (1985al

- 144 shy

On the supply side data were collected and estimates made of forest cover stocks yields and sustainable and accessible supplies of woodfuels Some sununary data on forest areas are given in Table 59 Table 510 presents sununary data on sustainable and accessible woodfuel supplies for present conditions as well as present woodfuel demand Woodfuel deficits and surpluses are shown for each region

Table 59 Contiguous Forest Cover by PrOVince Madagascar 1983-84

Faritany Natural Forest Plantations Forest Cover

( of far Itany)

Antananarivo Antsiranana Fianaranrsoa Mahajanga Toamasina Tollara

1145 15043 I 2850 21274 28137 44620

609 55

77 6 67

1021 119

29 34 13 14 41 27

Tota I 123069 2648 ~ 21

a Excludes the fanalamanga pine plantations Source UNDPWorld Bank [1985al

Although Table 510 shows that the country as a whole had surplus supplies on a sustainable basis it clearly identifies a major deficit for the Antananarivo district Further studies therefore focused on this area and the implications of introducing a range of new biofuel supply options The latter included rural afforestation and peri-urban plantations for fuelwood and charcoal the use of logging and sawmill residues for charcoal and the briquetting of charcoal fines or wastes and the briquetting of agricultural residues Also included were the upgrading of existing supply systems such as traditional charcoal production methods and tree coppicing for charcoal

Table 510 Woodfuel Demand and Supply Balance by Region Madagascar 1985 (thousand tonnes woodfuel equivalent)

Accessible SupplyDemand Faritany Sustainable Demand Deficit or (District) Supply Firewood Charcoal Total (Surplus)

Antananarivo 371 1287 887 27174 1803 Fianaranisoa 929 1123 300 1423 494 Antsiranana 688 231 92 323 (363) Mahajanga 1143 337 93 430 (713) Toamasina 1673 492 105 597 (1076) Tol iary 1946 464 83 547 (1399)

TOTAL 6750 3934 1560 5494 (1256)

Note Surpluses cannot be credited or transferred to deficit areas due to lack of transport infrastructure and high costs

Source UNDPWorld Bank [1985al

- 145 shy

A summary of the main findings is presented in Table 511 The calculation method is straightforward and can be followed easily by running down the rows of the table

On the demand side (Section A) rural and urban population and population growth rates are estimated separately as are per capita rural and urban household demand These are held constant A second analysis could have explored possible changes in per capita consumption and their effects on supply options Total demand is then calculated for each year

The second block of data (Section B) estimates the present sustainable woodfuel supply and holds this constant An alternative projection might have considered the effects of agricultural land changes on these supplies The contribution from modern fuels and from the increase of urban trees and woody residues is then added to these suppl ies to give a projection of the woodfue1 deficit with no intervention

The third block of data (Section C) sets out the increases in woodfuel supply from a range of proposed interventions (Le projects) designed to introduce new sources of biofuels upgrade existing resources and expand the supply and use of modern fuels Finally in Section 0 the supplies are totalled and an overall projection of woodfuel deficits is obtained

Supplementary tables not shown here could provide indications of the scale of the proposed interventions such as the areas of perishyurban plantations and number of seedlings required in each period

The penultimate step is to cost the various new supply options (and demand management options if these are included) This step is not shown here since it involves conventional and familiar methods Finally alternatives can be examined to provide one or more least cost set of options which can be compared for their effects on supplydemand deficits and balances

It is this final comparison with its presentation of associated costs and indications of the scale of interventions required that will attract the most attention from local officials aid agencies and others indeed that will form the starting point for negotiations on project selection and detailed project design possibly leading to eventual project implementation

However it cannot be stressed strongly enough that the paper assessments described above are only a starting point for a more practical and meaningful energy strategy or set of projects

Taple ll Projected Supply-Demand Balance for Household Energy Antananarivo Madagascar (thousand tons of wood equivalent twe)

198] 1985 1987 1989 1991 993 995

Urban Population (000) 69 5 7623 8405 92fj6 02 6 1263 2417 A I Rural Population (000) 2845 2304 24302 25632 27034 28514 30074

Total Population (1000) 28760 30664 32706 34898 37250 39717 42492 Total Energy Demand (1000 twe) 21114 22704 24206 2581 27526 29360 3320

Sustainable Supply Antananarivo Farltany

From Plantation (1000 twe) 32992 317 38 30533 29376 28264 27197 26172 From Forests (000 twe) 4582 4582 4582 4582 4582 4582 4582

Toamaslna Faritany From Plantation (000 twe) 12960 2960 2960 2960 2960 12960 12960 From Forests (000 twe) 28151 28151 28151 28151 28151 28151 28151

B I Total Sustainable Supply (000 twe) 7869 7143 7623 7507 7396 7289 7187 Existing Modern Fuels

Electricity (000 twe) 91 100 111 122 134 148 63 LPG (000 twe) 624 688 759 837 922 107 121 Kerosene (000 twe) 97 07 18 130 144 158 175 Sub-total (000 twe) 812 896 988 089 200 1323 1459

Urban Trees and Woody Residues (000 twe) 633 681 726 714 826 88 940 Deficit without Intervention (000 twe) 800 3384 4870 1644 18104 19866 2 735 CJ Deficit In ha equivalent (000 ha plantation) 983 1115 1239 1370 1509 1656 1811

New Sources Charcoal

Haut Mangoro Pine 00 00 187 187 187 87 87 Logging Residues 00 00 323 573 1020 813 3225

CI Sawm I I I Wastes 00 00 21 37 65 15 205 Lac Aloatra Charcoal Briquettes 00 00 00 00 39 112 228

Tota I Charcoa I 00 00 530 797 1311 2228 3846 Agricultural Residues Rice Husk Briquettes 00 00 35 63 11 198 -352

Sub-Total A 00 00 530 797 13 2228 3A46 to J

Upgraded Production o I Traditional Charcoal 00 00 217 433 650 866 1085

CoP ice Management 00 00 32 58 02 182 324 Sub-Total B 00 00 249 49 752 1049 407

Ex~anded Modern Fuel Sup~l~ Kerosene 00 00 89 158 281 500 890

E I Electricity 00 00 155 303 594 105 Sub-Total C 00 00 89 313 585 1095 1995

Total Supply 9314 9320 10240 1034 2181 4062 784 Deficit 11800 13384 13966 14717 5345 15297 14135

UNOPAlorld Bank 11985al~

- 147 shy

Annex 1

TYPICAL ENERGY CONTENT OF FOSSil AND BIOMASS FUELS

Moisture Content Typical Sol id Fuels Wet Basis Net Heating Values I

( mcwb) (MJkg)

Biomass Fuels

Wood (wet freshing cut) Wood (air-dry humid zone) Wood (air-dry dry zone) Wood (oven-dry) Charcoal Bagasse (wet) Bagasse (air-dry) Coffee husks Ricehulls (air-dry) Wheat straw Maize (stalk) Maize (cobs) Cotton gin trash Cotton stalk Coconut husks Coconut shells Dung Cakes (dried)

Fossil-Fuels

Anthrac ite Bituminous coal Sub-bituminous coal

lignite Peat

lignite briquettes Coke briquettes Peat briquettes

Coke

Petroleum coke

40 20 15 0 5

50 13 12 9

12 12 11 24 12 40 13 12

5 5 5

10-9

155 66

200 290 82

162 160 144 152 147 154 119 164 98

179 120

31~4

293 188

113 146

201 239 218

285

352

- 148 shy

TYPICAL ENERGY CONTENT OF FOSSIL AND BIOMASS FUELS (continued)

Specific Li qu I d Fuel s Gravity Net Heating Values

(MJkg) (t-tJ1 itre)

Fossil Fuels

Crude 01 I 086 419 367

LPG 054 456 246 Propane 051 457 233 Butane 058 453 263

Gasol ine 074 439 326 Avgas 071 443 315 Motor gaso I I ne 074 440 326 Wide-cut 076 437 333

White spirit 078 435 340

Kerosene 081 432 350 Aviation turbine fuel 082 431 354

Disti I late fuel oil Heating 01 I 083 430 357 Autodiesel 084 428 360 Heavy diesel 088 424 373

Residual fuel 01 I 094 415 390 Light 093 418 389 Heavy 096 414 398

Lubricating oils 0881 424 373 Asphalt 105 370 389 Tar 120 385 463 Liqui fied natural 042 528 222

gas

Biomass-Derived liquids E1hanol 079 276 219 Methanol 080 209 168

- 149 shy

Gas Net Heating Value

(MJm3)

Fossil Fuels Natural Gas 348

Refinery Gas 461

Methane 335 Ethane 595 Propane (LPG) 858 Butane (LPG) 1118

Pentane 1340 Coke oven gas 17 6 Town gas 167

Biomass-Derived Producer gas 59 Digester or Biogas 225

Electricity 36 MJkWh

~ Based on given moisture contents

Note For biomass fuels these data should be used only as rough approximations

Sources Biomass fuels--various (see text) modernnon-traditional fuels--FEA (1977)

- 150 shy

Annex 2

PREFIXES t UNITS AND SYMBOLS

I Prefixes and Symbols

SI American

thousand 103 k kilo M million 106 M mega KK billion 109 G giga G

1012trillion T tera T 1015quadrillion P peta

II EnerSI Symbols

SI

J joule Wb Watt-hour

AmericanGeneral

cal kcal calorie kilocalorie (103 cal) Btu BTU British Thermal Unit

Q Quadrillion Btu or Quad (1015 Btu)

toe TOE Metric tons of (crude) oil equivalent (defined as 107 kcal--41868 GJ in statistics employing net heating values)

tce TeE Metric tons of coal quivalent (defined as 07 x 10 kcal--293l GJ in statistics employing net heating values)

twe Thousand tons of wood equivalent

boe BOE Barrels of (crude) oil equivalent (approx 58 GJ)

bbl BBL Barrels of oil (crude or products) (equals 42 US gallons)

Note American and SI systems use M differently

- 151 shy

PREFIXES UNITS AND SYMBOLS (continued)

III Power (and Electricity) Symbols

W v V a A

kVA

BTUhr hp

bd bId bdoe

IV Weights and Measures

g kg lb lbs

t te ton lt ton st ton

tpa tpy

m km mi

2sq m mha ac

1 3cu m m

gal

SCF CF

V Biomass amp Other

od 00 odt ODT

ad AD mcwb mcdb

MAl GHV NHV

SI

Watt Volt Ampere kilovolt-ampere

AmericanGeneral

British Thermal Units per hour Horsepower Barrels of oil per day Barrels of oil equivalent per day (Barrels of daily oil equivalent)

Gram or gramme kilogram Pound pounds Metric tonne or 106 g (SI) Long ton (Imperial 2240 pounds) Short ton (US 2000 pounds) Tons per year

Meter kilometer (SI) Miles

Square metel Hectare (10 m2) Acre

Liter litre (SI) Cubic meter gallon (US or Imperial)

Standard cubic foot (used for gases at normal temperature and pressure)

Oven dry Oven dry ton Air dry Moisture content wet basis Moisture content dry basis Mean Annual fncrememt Gross and Net Heating Value

CONVERS ION FACTORS (con tinued)

VOLUME To convert ---) 3 It3 yd3 UK I I oz UK pt UK gal US I I oz US pt US gal

2

cubic metre 1 10000 -3 28317 -2 76455 -1 28413 -5 56826 -4 45461 -3 29574 -5 47318 -4 37854 -3 itre 99997 +2 1 28316 +1 76453 +2 28412 -2 56825 -1 45460 0 29573 -2 47316 -1 37853 0

cubic foot 35315 +1 35316 -2 1 27000 +1 10034 -3 20068 -2 16054 -I 10444 -3 16710 -2 13368 -1 cubic yard 13080 0 13080 -3 37037 -2 1 37163 -5 74326 -4 59461 -3 38681 -5 61889 -4 49511 -3 UK fluid ounce 35195 +4 35196 +1 99661 _2 26909 +4 20000 +1 16000 +2 10408 0 16653 _I 13323 +2 UK pi nt 17598 +3 17598 0 49831 +1 13454 +3 50000 -2 1 80000 0 52042 -2 83267 -1 66614 0 UK gallon 21997 +2 21998 -I 62286 0 16816 +2 62500 -3 12500 -I 65053 -3 10408 0 83267 -1 US fluid ounce 33814 +4 33815 +1 95751 +2 25853 +4 96076 -1 19215 +1 15372 +2 1 16000 +1 12800 +2 US pi nt

US gallon

21134 26417

+3 +2

21134 26418

0 -1

59844 74805

+1 0

16158 20197

+3 +2

60047 75059

-2 -3

12009 15012

0 -1

960761 12009

0 0

62500 78125

-2 -3

I 12500 -1

80000 0 w

CONVERSION FACTORS (continued)

MASS To conllert---gt kg t Ib UK ton sh ton

Into kilogram tonne pound UK ton (=Iong ton) short ton

10000 22046 98421 11023

-3 0

-4 -3

10000 1

22046 98421 11023

+3

+3 -1 0

45359 45359

44643 50000

-1 -4

-4 -4

10160 10160 22400

11200

+3 0

+3

0

90718 90718 20000 89286

+2 -1 +3 -1

WORK ENERGY HEAT To Convert---gt J kcal kWh hph Btu

Into joule 1 41868 +3 36000 +6 26845 +6 10551 +3 ki localorle 23885 -4 1 85859 +2 64119 +2 25200 -1 k i lowatt hour horsepower hour

27778 37251

-7 -7

11630 15596

-3 -3 13410 0

74570 -1 29307 39301

-4 -4

U1 po

British Thermal unit 94782 -4 39683 0 34121 +3 25444 +3

POWER ENERGY CONSUMPTION RATE convert---gt W kW CV kcal min Btu mln- 1

Into watt ki lowatt metriC horsepower

(cheval-vapeur) horsepower ki localorie per minute British thermal unit

per minute

10000 13596

13410 14331

56869

-3 -3

-3 -2

-2

10000

13596

13410 14331

56869

+3

0

0 +1

+1

73550 73550

98632 10540

41827

+2 -1

-1 +1

+1

74570 74570 10139

1 10686

42407

+2 -1 0

+1

69780 69780 94874

93577

39683

+1 -2 -2

-2

0

17584 17584 23908

23581 25200

+1 -2 -2

-2 -1

Note A few examples 2 yd = 2 x 49374 international nautical miles

x 10 -4

1 acre = 40469 x 10 3 square meters

3 mile 2 = 3 x 40145 x 109 square inch

- 155 shy

Annex 4

GLOSSARY

Air-dried weight

Anaerobic processes

Bagasse

Biogas

Biomass fuels

British Thermal Unit (BTU)

Calorie

Coal equivalent

A fuels moisture content after being exposed over time to local atmosshypheric conditions

A name for some biomass digestion systems these are biological chemical processes which typically break down organic material into gaseous fuels in the absence of oxygen

The burnable fibre remaining after sugar has been extracted from sugar cane

A gas of medium energy value (22HJm3) generally containing 55-65 methane and produced by anaerobic decomposition of organic materials such as animal wastes and crop residues

Combustible andor fermentable organic material for example wood charcoal bagasse cereal stalks rice husks and animal wastes

A measure of energy specifically the heat required to raise the temperature of one pound of water by one degree Fahrenheit

A metric measure of energy specifically the heat required to raise the temperature of one gram of water from 145deg to I55degC at a constant pressure of one atmosshyphere

The heat content of a fuel in terms of the equivalent heat contained in an average ton of coal Measures for local coal or international standards may be used

- 156 shy

Coal replacement

Commercial energyfuel

Conventional energyfuel

Combustion efficiency

Energy content as received

Energy content of fuel at harvest

Gross Heating Value (GHV)

A measure of the amount of coal that would be needed to substitute for other fuels in an energy conversion process

This term is often used in the context of developing countries to refer to all non-traditional or nonshybiomass fuels such as coal oil natural gas and electricity Commercialized (or monetized) energy includes traditional fuels that are exchanged for cash payments

Another term for commercial energy as defined above

The utilized heat output of a combustion technology divided by the heat content of the fuel input See Chapter II for other definitions and equations

The energy content of a fuel just before combustion It reflects moisture content losses due to airshydrying or processing (eg kiln or crack drying logging or chopping) For these reasons the energy content as received is generally higher per unit weight than that of the fuel at harvest

Normally used for biomass resources the energy content of a fuel at the time of harvest It is often referred to as the green energy content

This is the total heat energy content of a fuel It equals the heat released by complete combustion under conditions of constant volume (i e in a bomb calorimeter) It equals the thermodynamic enthalpy of the fuel and depends only on the fuels chemical composition and weight which includes contained water It is sometimes referred to as the higher heating value

- 157 shy

Moisture content dry basis (mcdb) The ratio of the water weight of a fuel to the oven-dry (solid fuel) weight expressed as a percentage

Moisture content wet basis (mcwb) The ratio of the water weight of a fuel to the total (water plus solid fuel) weight expressed as a percentage

Net Heating Value (NHV) This is a practical measure of the heat obtained by complete combustion of a fuel under the usual conditions of constant pressure It is less than the Gross Heating Value by an amount representing mainly the chemical energy and latent heat involved in vaporization of exhaust gases and water vapour etc It is sometimes referred to as the lower heating value

Oven-dried weight The weight of a fuel or biomass material with zero moisture content

Photovo1taic (PV) cell Solid state technology which converts solar energy directly into electricity

System efficiency System efficiency in the context of this handbook is the total efficiency of converting primary energy resources into utilized energy

Traditional energyfuel In the context of developing countries firewood charcoal crop residues and animal wastes or other biomass fuels See Commercial EnergyFuel Conventional Energy Fuel

- 158 shy

Utilized energy

Green weight

The energy actually utilized for a specific task such as cooking or lighting Energy losses in conversion technologies ensure that utilized energy is always less than energy as received

The weight of a biomass fuel at harvest including moisture content

Hote Definitions come primarily from the text but some are adopted from Renewable Energy Resources in Developing Countries World Bank January 1981

Annex 5

SUMMARY Of CLASSES OF CONSTRAINTS FOR WOOD STOVE DESIGNS

CLASS Material

ADVANTAGES DISADVANTAGES SOLUT ION OPT IONS

Clay (I) available in more abundance non-uniform in quality will require beneficiation

(II) fabrications do not need sophisticated machinery

quality control difficult

(iii) runs cool stable on the ground and safe in operation

heavy not portable to be built In-situ not amenable to marketing through conventional channels uncershytain life expectancy

Ceramic (I) same as with clay

(Ii) quality control better than with clay

(III) lighter portable and can be marketed more easily

material requirement more stringent special kilns required

runs hotter than clay rather high risks of shattering amp uncertain life expectancy

(i) clay with metal reinforcements

(Ii) clay with ceramic inner liner

(ill) metal with clayceramic inner liner

Jl 0

Metal (I) available according to designers desires

(Ii) excellent quality control posslbl I Itles

not as accessible as clay --most of these Improvements cost more but overcome many disshyadvantages of the individual sophisticated machinery for fabrishycation dependent on the material for example thick steel sheet requires special Welding and bending equipment

(Iii) light portable and excellent marketability

runs hot special features for stability required

CLASS ADVANTAGES DiSADVANTAGES SOLUTION OPTIONS Manufacturing Method

Owner-bu i It

tinerant art isan

Industrial

(i) little or no cash outlay

(Ii) small design changes to accommodate Individual variations

(iii) individual independence

(i) skilled craftsmanship at work quality control better

(Ii) possible to bring in new Ideas of design with time

(iii) promotes the formation of a guild of artisans slight movement towards a monetized economy

(i) a standard product with a reliable performance possible

(I i) could sustain an In-house design capability for continshyuous product innovation

(iii) sophisticated marketing techniques feasible

(Iv) helps In moving subsistence living patterns into producshytive entreprenurlai patterns

Poor quality contrOl material procurement difficult significant design changes difficult

no speCial community advantage maintains subsistence existence

labor of craftsman needs to be paid for entity responsible for RampD design and marketing isolated work situation with no stimulus for radically new ideas

required to adjust to the artisans method and time of work

requires higher capital outlay and sophisticated infrastructure--both unavailable now in rural areas

product may not be avai lable for the really poor

(not connected with design manufacshyturing but with organization) (i) a single large unit manufacturing elements like grates top plates and chimneys servicing a large number of Itinerant artisans (ii) several small scale production units operated by a single management

I- 0shyo

CLASS ADVANTAGES DISADVANTAGES SOLUTION OPTIONS Design Type

Two-hole (I) higher thermodynomlc poor flexibility in operation single point efficiency firing heavy structure better to work with both designs system not amenable to conventional let the users decide

marketing approach

Single pay (i) great flexibility for the lesser thermodynamic efficiency operator

(I i) lighter structure (i ii) easily marketable

t- 0 t-

- 162 shy

Annex 6

PROCEDURES FOR TESTING STOVE PERFORMANCE

Efficiency testing procedures must be standardized so that results can be compared Procedures and results must also be reproducible and well documented Furthermore efficiency tests should take into account the cooking practices of a given region or country Since these factors vary widely the requirements for measuring stove efficiency often can conflict To resolve this problem three separate test procedures have been established the Water Boiling Test (WBT) Controlled Cooking Test (CCT) and Kitchen Performance Test (KPT) The set of Provisional International Standards for testing the efficiency of wood-burning cookstoves was developed at a VITA conference in 1982 with the involvement of the major ICS programs

The three tests basically cover the spectrum from highly controlled easily measured tests (WBT) to more realistic but consequently more variable test procedures (KPT) The WBT measures efficiencies at the high power phase when water is brought to the boil and the low power phase when water is kept simmering just below boiling In the WBT measurements of efficiencies at maximum power (p ex) and minimum power (Pmin) phases are taken and an average efflciency calculated Using an average efficiency is important since stove efficiency may actually drop to near zero during the simmering low power phase These power ranges reflect common cooking requirements in developing countries where water is often brought to a rapid boil for cooking rice or other cereals and then simmered for long periods

WBT test results should give reliable comparisons so long as the procedures are not varied and are well documented Consistency in seemingly minor matters such as using or not using a lid the type of pot and fire maintenance are important to the results

Although WBT results give efficiencies which are easily comparable they may not reflect efficiencies achieved when cooking a meal The Controlled Cooking Test was developed to allow for this In the CCT a regular meal representative of a region or country is cooked by a trained worker to simulate actual cooking procedures followed by local households Cooking efficiencies derived from these tests should correspond more closely to actual household efficiencies As with the WBT these tests are conducted in a laboratory or in the field by trained stove technicians or extension workers Given the many variables in the CCT that could affect efficiency results these tests require careful measurement of ingredients and documentation of pot sizes pot types fuel and sequencing of procedures by the cooker

- 163 shy

The IPT is a more realistic and even more specific test than the CCT Using individual families under normal household conditions household cooks prepare their usual meals with the improved stove These tests show the impact of a new stove on the overall household energy use IPT testers may also demonstrate to potential users the fuel saving quality of the new stove and recommend more efficient operating practices This test thus can be far more than a measure of stove efficiency by combining scientific data gathering with active household participation

- 164 shy

Annex 7

METHODS FOR ESTIMATING PAYBACK TIMES FOR STOVES

If the costs of operating stoves include repairs and periodic stove replacement mathematical expressions for estimating payback times are quite complex It is usually far simpler to use graphical methods

Figure Al shows the cumulative costs of an improved stove and the existing unit which it replaces plotted against time I is the initial cost of the new stove which is replaced once during the period shown 0 is the replacement cost of the existing (old) stove which is replaced-twice R denotes repair costs which may be different for the new and old stoves The slopes of the cost curves are given by the fuel cost per uni t of time ie by fuel consumption per unit of time multiplied by the fuel price

The payback time can be read off the plot at the point where the cost curves intersect

More sophisticated analyses can be made in which the initial and repair costs are discounted using an appropriate rate (eg the prevailing interest rate on capital borrowing) This sophistication is rarely justified for small investments such as stoves especially given the large uncertainties over costs lifetime between repairs or fuel savings

FIGURE A1 Estimating Payback Times

Cost I I I I I I I I I

r I Payback Period

~-~ Time

World Bonk-307365

- 165 shy

If the costs and timing of repairs are unknown a good approxillation to the payback time can be made simply by equating the investment plus fuel costs of the new stove to the fuel costs of the old unit for any time period thus

I + F x P = f x p

Where I is the investment cost of the new stove F f are the quantities of fuel consumed per unit of time (day week etcgt by the new and old stove and 2 represents fuel prices The payback period in the time units used for ~ ~ is given by

Payback period = I I (f x p F x p)

- 166 shy

Annex 8

IMPACT OF URBAN WOODFUEL SUPPLIES

The supply of urban woodfuels is almost exclusively on a commercial basis In small towns woodfuel supply mechanisms tend to be relatively informal Rural suppliers may themselves transport fuel to the towns using donkeys or bullock carts carrying it on buses or bringing it in by headload Some sell to dealers while others trade directly in the market place

In larger cities trade is more often organized around a series of wholesale depots from which smaller retailers obtain their supplies Wood and charcoal are usually brought in by truck from the surrounding areas

The Kenyan charcoal market is to a large extent controlled by truck owners They purchase the charcoal from rural producers and sell it through their own outlets in the cities In some cases charcoal is picked up on the way back from delivering other goods to outlying districts This alters the economics completely and opens up a much wider area of potential sources As a result charcoal may sometimes be brought from surprisingly long distances away Some of the trucks carrying charcoal to Nairobi come from as far away as the Sudanese border 600 kilometers to the north

As trucks and other vehicles are usually the predominant method of transporting woodfuel supplies to urban areas the road network has a major bearing on the sources of supply The opening up of forest areas to logging for example often results in the development of a concomitant trade in woodfuel Simply improving a road into a village so that it can be used by a bus may have the same effect

As long as rural areas remain relatively isolated the effects of increasing woodfuel pressure usually will be gradual When areas become subject to concentrated urban demands however this can bring about a dramatic increase in the depletion rate The cash incentive created by these demands means that people have a much stronger motive to cut trees They will go further afield to gather wood and will take greater risks in entering and illegally cutting trees from forests and unprotected private lands

The impact of an urban woodfuel market has been described as follows

Note Extracted with permission from Barnard [1985]

- 167 shy

(it) creates not only a distinctive spatial character for fuelwood production bullbullbullbut also changes the character of fuelwood exploitation It is more selective of tree species whether for charcoal production or urban fuelwood for consumers and it is also more wasteful of the wood resource It employs paid labor sometimes specialized cutting or processing skills and it has to deal with problems of storage and seasonality in production and supply It also diverts wood fuel from subsistence use as poor people in areas of short supply sell their wood or charcoal to higher income groups in the towns [Morgan 1983]

In some countries wood cutting is carried out by large wellshyorganized gangs sometimes operating in collusion with local forestry officials so as to avoid cutting regulations and licence fees More often however it is the poor who are involved as families are forced to turn to wood sell ing because of the lack of other income earning opportunities The reasons behind this have been described with specific reference to Karnataka State in India

Denudation of forests has often been viewed merely as the result of rural energy consumption However for a villager who has no food the attack on forests is for collection of firewood for sale in urban and semi-urban centres rather than his own consumption because selling firewood is often the only means of subsistence for many poor families This firewood with the help of bus and truck drivers goes to the urban markets like Bangalore bullbullbullTheft of wood as a means of survival is becoming the only option left for more and more villagers Recently 200 villagers were caught stealing firewood in the Sakrabaile forest of Shimoga district and one person was killed in a police encounter [Shiva et ale 1981]

Trees on private land may also be sold in response to external commercial demands The amount of these sales will depend on the prices being offered and on the financial needs of the farmers who own them In poor areas or when harvests fail farmers are sometimes forced to cut their trees to earn cash In Tamil Nadu it has been observed in some vi11ages that

distress sale of trees because of drought conditions is reported This indicates that the villagers resort to short term exploitation of fuel resources in drought periods when their incomes fall drastically unmindful of the long term consequences of their act [Neelakantan et ale 1983)

The deforestation that has occurred around the city of Kano in Northern Nigeria over the last 25 years also illustrates this Formerly there was a tradition whereby farmers used to lop branches from the tree~ on their land during the dry season and transport them into the town on donkeys to sell in the market While in town they picked up dung and

- 168 shy

sweepings from the streets which they carried home and used as fertilizer on their fields With growing wood demands in the city the incentive to cut trees has increased As a result what was once a relatively stable system has broken down to the extent that farming land within a 40 kilometer radius of the city has been largely stripped of trees

Charcoal making for the urban market is also a major cause of tree depletion in some areas In the Sahel this has a long history The widespread destruction of acacia torti1is for example can be traced back to charcoal production carried out for the trans-Saharan camel trade [Cori110n and Gritzer 1983]

The opening up of river communications has also led to severe deforestation along the flood plain of the Senegal River where once extensive stands of Acacia ni10tica have been cut for charcoal production Elsewhere in the Sahel region improvements in road communications have resulted in similar destruction as urban charcoal markets become accessible to more remote rural areas [Coril10n and Gritzer 1983] In Kenya the provision of access roads to Mbere district has reportedly led to a substantial increase in the number of trees being felled for charcoal for urban markets with a total disappearance of large hardwoods such as Albizia tangankiensis [Brokensha Riley and Castro 1983]

The severe impact of cutting for charcoal has also been noted in a detailed study of the woodfue1 position in Haiti Charcoal production was found to be particularly destructive because live trees are harvested as opposed to the dead branches and twigs which provide the bulk of rural firewood supplies As is frequently the case charcoal production in Haiti is carried out only by the very poor The attitude of local people to the resulting deforestation was summarized as follows

Local residents 1n all of the research sites recognized deforestation as a great problem Deforestation is seen as contributing to floods and drought Even young adults can remember when the hillsides now denuded were covered with trees Furthermore charcoal production is perceived as the cause of this deforestation More to the point poverty is seen as the cause of deforestation because only poverty leads a person to make charcoal Rather than resentment against charcoal makers as destroying a natural resource there is great sympathy for such people [Conway 1979]

Urban woodfuel demand thus can be a major factor in causing deforestation in the area over which it extends It reinforces local demand and can greatly accelerate the depletion process It is therefore important that urban demands are distinguished from local demands when methods of countering the effects of woodfuel scarcities are being considered

- 169 shy

Annex 9

STAGES OF SOIL DEGREDATION DUE TO TREE LOSS AND REMOVAL OF CROP RESIDUES IN ETHIOPIA

At the rate at which peasant agriculturalists are currently clearing the fringes of natural high forest this resource will be lost in about 30 years As in the past during this first stage of forest clearing for the purpose of developing land for food production local fuel wood is abundant At present perhaps 20 mill ion cubic meters of wood the same quantity that is consumed in all the households of Ethiopia are burnt off during agricultural clearing each year It is only sometime later that trees begin to be harvested primarily for fuel Beyond this point it appears that a critical transition of decline begins within subsistence agriculture whereby the growing scarcity of woodfue1s is linked inextricably to falling crop and animal production This transition leads to and is clearly exacerbated by growing urbanization in Ethiopia as the nature and level of fuel use for household cooking for most urban dwellers closely resembles that for their rural counterparts The demand for woodfue1s and ultimately for any combustible residue by urban dwellers or members of any concentrated settlement without a sufficient independent resource base (ie state farms) becomes an intolerable burden on rural productivity A conceptualization of the perceived stages of this transition follows below and in Figure A2

Stage 1 The rate of timber harvested locally for all purposes (fuel construction tools fences) exceeds for the first time the average rate of production The existing timber resource is then progressively Itmined firewood remains the main fuel source Nutrient cycle No 1 begins to decline though with imperceptible impact on food production The general reason for the imbalance is population growth The specific reasons include urbanization and major land clearing (eg state-farms) whereby firewood and charcoal become cash crops leading to overcutting relative to purely local subsistence requirements

Stage II The great majority of timber produced on farms and on surrounding land is sold out to other rural and urban markets Peasants begin to use cereal straw and dung for fuel the relative proportions depend on the season Both nutrient cycles No 2 and No 3 are breached for the first time and nutrient cycling diminishes Combustion of crop residues and dung leads to lower inputs of soil organic matter poor soil structure low retention of available nutrients in the crop root zone and reduced protection

Note Quoted with permission from Newcombe [1985]

FIGURE A2 Pattern of Deterioration in Ethiopian Agroecosystems

Breach Dung Removed as

Fuelwood Substitute Breach Tree Cover Removed

for Firewood

o

Cycle No2 Grass amp Crop Residue

Nitrogen-Fixing amp Retention Mineral Retention amp Cycling

Spill Erosion of Nutrient amp

Humus Rich Topsoil as Main Nutrient Cycles

are Breached

BreaCh Overgrazing Scavenging for Fuelwood Substitute

World Bank-3073612

- 171 shy

from the erosive effect of heavy rainfall Hence topsoil nutrient reserves begin to decline (See spill in the Figure)

Stage III Almost all tree cover is removed Now a high proportion of cow dung produced is collected the woodier cereal stalks are systematically collected and stored and both are sold for cash to urban markets The yields of cereal crops and in consequence animal carrying capacity begins to decline Draft animal numbers and power output are reduced hence the area under crop also falls Soil erosion becomes serious Nutrient cycle No 1 ceases altogether

Stage IV Dung is the only source of fuel and has become a major cash crop All dung that can be collected is collected All crop residues are used for animal feed though they are not sufficient for the purpose Nutrient cycle No 2 is negligible and No 3 is greatly reduced Arable land and grazing land is bare most of the year Soil erosion is dramatic and nutrient-rich topsoil is much depleted Dung and dry matter production have fallen to a small proportion of previous levels In such a situation extended dry periods can be devastating because the ecosystem loses its capacity to recover quickly

Stage V There is a total collapse in organic matter production usually catalyzed by dry periods which were previously tolerable Peasants abandon their land in search of food and other subsistence needs Starvation is prevalent Animal populations are devastated Rural to urban migration swells city populations increasing demand on the rural areas for food and fuel and the impact of urban demand is felt deeper into the hinterland (the urban shadow effect)

This transition from the first to the final stage is in process right across Ethiopia and has reached the terminal phase in parts of Tigrai and Eritrea The only way to prevent the current situation in the rema1n1ng populous and fertile areas from sliding toward the terminal state of Stage V is to develop a strategy which will

(a) remove the dependency of urban settlements on their rural hinterlands for woody fuels and

(b) reestablish a dynamic equilibrium between supply and demand for firewood in rural areas

While the development of peri-urban fuelwood plantations is an obvious component of a strategy to serve the first objective the time required to do this is such that even if design work began inunediately the production of woodfuels would hardly begin to be augmented before the end of the decade Without urban self-sufficiency it will be extremely difficult to achieve the second objective as biomass fuels will continue to drain from the rural areas to the towns and cities In addition the

- 172 shy

situation of Northern Ethiopia where in many places agricultural ecoshysystems have deteriorated to stages IV and V demands special and possibly separate consideration because of the huge scale of the problem and the implied investment and the added complexity of local hostilities

- 173 shy

BIBLIOGRAPHY

Acharya M and L Bennett [1981] The Rural Women of Nepal an Aggregate Analysis and Summary of Eight Village Studies CEnAStatus of Women Project Vol 2 Field Studies Part 9

Aggarwal GC and NT Singh [1984J Energy and Economic Returns from Cattle Dung as Manure and Fuel Energy 9 87

Aiyasamy PK et a1 [1982J Social Forestry Project in Tamil Nadu Survey Report No2 Tamil Nadu Agricultural University Coimbatore India

Alam M et a (1983] Fuelwood Survey of Hyderabad Osmania University Hyderabad amp Resources for the Future Washington DC

Anderson D and R Fishwick (1984J Fuelwood Consumption and Deforestation in African Countries World Bank Staff Working Paper 704 Washington DC

Anon (1976J Le deboisement en Haute-Volta besoins en bois de chauffe de Ougadougou Le Developpement Voltaique 40

Anon [1981aJ Balance Energetico Nacional 1970-80 Vol 2 Comision Nacional de Energia Panama

Anon [1981b] Policy Issues in Ethiopian Energy Development Stockholm Beijer Institute

Anon (1982J Energy Mission Report Cook Island Resource Systems Institute East-West Center Hawaii

Anon (1983J Uso Efficiente de Energia en Chile Chilean National Committee for the World Energy Conference Santiago Chile

Arnold J and R deLucia (1982J Rural Energy Surveys an Experience from Thailand Meta Systems Inc Boston Mass

Aulaqi NA [1982] Household Energy and Tree Seedling Demand Survey in the Southern UplandS of the Yemen Arab Republic Sanala University Yemen

Ban A [1985J Boiling Point Intermediate Technology Development Group Stoves Project ITDG Applied Research Section Reading UK

- 174 -

Bajracharya D [1981] Implications of Fuel and Food Needs for Deforestation an Energy Study in a Hill Village Panchayat of Eastern Nepal PhD dissertation Sussex University England

Baldwin S and S Strasfogel [1983] Compte rendu de test sur foyers metalliques et Proposition - Hali CILS Bamako Hali

Barnard G [1985] Woodfuel in Developing Countries London EarthscanInternational Institute for Environment and Development

Barnard G and L Kristofferson [1985] A ricultural Residues as Fuel in the Third World London Earthscan International Institute for Environment and Development Stockholm Beijer Institute

Bene JG HW Beall and HB Marshall [1978] Energy from Biomass for Developing Countries Ottawa International Development Research Center

Bernardini TO [1983] Rural-Urban Household Energy Survey Ministry of Mines and Energy Addis Ababa Ethiopia Italy Centro Studi Energia (CESEN) Genova Italy

Best M [1979J The Scarcity of Domestic Energy A Study in Three Villages Paper No 27 South African Labour and Development Research Unit Cape Town South Africa

Bhattarai TW K Shrestha and J Gabriel Campbell [1984] Wood Burning Stove Programme of the Community Forestry Project in Nepal (a Case Study) FAOUNDPNepa1 Department of Forests

Bialy J [1979] Firewood Use in a Sri Lankan Village a Preliminary Survey School of Engineering Science University of Edinburgh Scotland

[1983] Energy Flows in Subsistence Agriculture A Study of a Dry Zone Village in Sri Lanka PhD dissertation School of Engineering University of Edinburgh Scotland

Bowonder B and K Ravishankar [1984J Energy Use in Rural Areas of India Processed Administrative Staff College of India Hyderabad India Processed

Bowonder B et ale [1986J Energy Use Patterns in 10 Villages 1t

Administrative Staff College of India Hyderabad India

Briscoe J [1979] Energy Use and Social Structure in a Bangladesh Village Population and Development Review 5615-41

- 115 -

Brokensha D BW Riley and AP Castro [1983J Fuelwood Use in Rural Kenya Impacts of Deforestation New York Institute for Development Anthropology

CBC [1985J Report on Consumer Finances and Socio-economic Survey 198182 Sri Lanka Central Bank of Ceylon Colombo sri Lanka

CBS [1980J The RuralUrban Household Energy Consumption Survey October-November 1918 Nairobi Central Bureau of Statistics

Cecelski E [19841 The Rural Energy Crisis Womens Work and Family Welfare Perspectives and Approaches for Action 1t

Rural Employment Policy Research Programme International Labour Organisation Geneva Switzerland

Cline-Cole R [1981] ItFirewood in a Rural Settlement in Sierra Leone It In Anon [1981a]

Conway FJ [1919] A Study of the Fuelwood Situation in Haiti 1t Haiti Mission US Agency for International Development

Corillon C and JA Gritzer [1983J Environmental Change in the West African Sahel Washington DC National Research CouncilNational Academy Press

DCS [1983] ItLabour Force and Socio-economic Survey 198081 Sri Lanka Department of Census and Statistics Colombo Sri Lanka

de Lepeliere G [1982)= A Short Reflexion on Woodburning Cooking Stove Performances Efficiencies and Fuel Savings In KK Prasad and E Sangan eds Technical Aspects of Woodburning Cookstoves Eindhoven The Eindhoven University of Techshynology The Netherlands

de Lepeliere G et ale [1981] A Woodstove Compendium Eindhoven The Eindhoven University of Technology The Netherlands

de Montalembert MR and J Clement [1983] Fuelwood Supplies in the Developing Countries Forestry Paper No 42 Rome FAO

Diaz F and A del Valle [1984] Energy Sources and Uses in the Poor Rural Sector of Chile - Synthesis of Eight Case Studies 1t Centro de Investigacion y Planificacion del Medio Ambiente (CIPMA) Santiago Chile Processed

- 176 -

Doat J and G Petroff [1975J liThe Carbonization of Tropical Woods Laboratory Tests and Industrial Prospects The Journal of Tropical Wood and Forests (Reune Boiset Forets des Tropiques) 159 (Jan-Feb) 55-68

DOE [1983J Family Expenditure Survey 1982 11 Department of Employment London England

Douglas J [1981] Consumption and Supply of Wood and Bamboo in Bangladesh 1I UNDPFAOPlanning Commission (Govt of Bangladesh) Project BGD78010

Down S [1983J Household Energy Consumption in West Sumatra Implications for Policy Makers 1I Energy 8 (11)

EIA [1983] Residential Energy Consumption Survey Consumption and Expenditures April 1981 through March 1982 Energy Information Administration Washington DC

Evans M [19841 IIFirewood versus Alternatives Domestic Fuel in Mexico Commonwealth Forestry Institute Oxford England

FAO [1983a1 Forest Product Prices 1963-82 Rome

[1983bJ Wood and Energy FAO Forestry Department Tropical Report No1 Rome

[1983c] Wood Fuel Surveys Forestry for Local Community Development Programme GCpINT365SWE Rome

FBS [1983] Household Income and Expenditure Survey 1979 Karachi Pakistan Federal Bureau of Statistics Government of Pakistan

FEA [1977] Energy Interrelationships a Handbook of Tables and Conversion Factors for Combining and Comparing International Energy Data Washington DC Department of Energy Federal Energy Agency Report No FEAB-88166

Fleuret PC and AK Fleuret [19771 IIFuelwood Use at Lushoto Department of Anthropology University of California Santa Barbara Processed

[19781 Fuelwood in a Peasant Community a Tanzanian Case Study Journal of Developing Areas 12315-22

FMT [1983J National Transport Survey and Projections Lagos Nigeria Federal Ministry of Transport Report by Louis Berger International Inc New Jersey

- 177 -

Foley G [19861 Charcoal Making in Developing Countries London EarthscanInternational Institute for Environment and Development

Foley Gerald and G Barnard [1984] Farm and Community Forestry London EarthscanInternational Institute for Environment and Development

Foley Gerald and Patricia Moss [1983] Improved Cooking Stoves in Developing Countries London EarthscanInternational Institute for Environment and Development

French D [1981] Malawi Rural Energy Survey Energy Unit Ministry of Agriculture Lilongwe Malawi

[1985J The Economics of Reforestation in Developing Countries Biomass Energy Systems Building Blocks for Sustainable Agriculture Washington DC World Resources Institute

Furness CK [1981J Some Aspects of Fuelwood Usage and Consumption in African Rural and Urban Areas in Zimbabwe South African Forestry Journal 10-12

Geller HS and GS Dutt [1983] Measuring Cooking Fuel Economy In FAO 1983c Energy Research Group IDRC Ottawa and UN University Tokyo Japan

Gill J [1983] Fuelwood and Stoves Lessons from Zimbabwe Proceedings of the Indian Academy of Sciences 1 (March 1983) 79-94

Goldemberg J et a1 [1984] Energy Use in Brazil - A Study on End-use Energy Strategies Companhia Energetica de Sao Paulo

Gowen M [19851 Renewable Energy Assessments An Energy Planners Manual Honolulu The East-West Center

Grut M [19711 Developpment des res sources forestieres Togo marche de charbon de bois de bois de chauffage et de bois rondo Rome FAO

Gupta CL and KU Rao [1980] Domestic Energy Consumption in India (Pondicherry Region) Energy 5 1213-22

Guzman o [1982] Case Study on Mexico In MS Wionczek G Foley and A van Buren Energy in the Transition from Rural Subsistence Westview Press

- 178 shy

Hernandez J [19801 Household Energy Use in Non-OPEC Developing Countries Rand Corporation Report R-2515-DOE for US Department of Energy

Ho GE [19831 Crop Residues How Much Can be Safely Harvested School of Environmental and Life Sciences Murdoch University Australia

Hoskins MW [19821 Benefits Foregone as a Major Issue for FLCD Success Paper to Community Forestry Workshop USAID Washington DC

Howes Michael [19851 Rural Energy Surveys in the Third World A Critical Review of Issues and Methods IDRC Canada

Hughart David [1979] Prospects for Traditional and Non-Coventiona1 Energy Sources in Developing Countries World Bank Staff Working Paper 346 Washington DC

Hyman Eric [1986] liThe Economics of Improved Charcoal Stoves in Kenya Energy Policy 14

ICAR [1984] Annual Report 1982-83 All India Coordinated Research Project on Long Term Fertilizer Experiments New Delhi Indian Council of Agricultural Research

lEE [1980] Facts and Figures of Energy Consumption in the Residential and Commercial Sectors TOkyo Institute of Energy Economics

ILO [1979] Household Income and Ex enditure Statistics No3 (1968-76 International Labour Organisation Geneva Switzerland

Ishengoma RC [1982] Domestic Cooking with Charcoal in Morogoro Town University of Dar es Salaam and IDRC Charcoal Stove Project Cited in Cecelski [1984J

Islam N et a [1984] Study on Fue1wood and Charcoal in Thailand Bangkok Asian Institute of Technology and Brussels Development Directorate European Commission

Jones J and A Otaro1a [198t] Diagonstico Socio-Economico sobre el Consumo y Production de Lena en Fincas Pequenas de Nicaragua Turrialba Costa Rica CATIE

Joseph Stephen and Philip Hassrick [1984] Burning Issues 1m 1ementin Pilot Stove Pro rammes A Guide for Eastern Africa New York UNICEF Intermediate Technology Publications

- 179 -

Kamweti DM [1984] Fuelwood in Eastern Africa Present Situation and Future Prospects Rome United Nations Food and Agriculture Organisation (FO MISC8421)

Kaupp A and JR Goss [1981] State of the Art for Small-Scale Gas Producer Engine Systems Department of Agricultural Engineering University of California Davis

Kjellstrom B [1980] Producer Gas 1980 Stockholm Beijer Institute

Koenig D [1984] Energy Use in the Regions of Niori and Bougouni 1t Draft Report US Agency for International Development Mali Renewable Energy Project Cited in Cecelski [1984]

Kravis I (1982) World Product and Income International Comparisons of Real Gross Domestic Product Baltimore Md Johns Hopkins University Press

Kuyper J and W Mellink [1983] Supply Demand and Consumption of Fuelwood for Rural Households in Central Java1t Wageningen Holland Agricultural State University

Lal R [1976] ItSoil Erosion Problems of an Alfisol in Western Nigeria and its Control Advances in Agronomy 37 183

Lambert I [1984] Energy Needs in West African Countries Possible Contributions of Renewable Energy Energy Research Group IDRC Ottawa and UN University Tokyo Japan

Leach G et ale [1986] Energy and Growth a Comparison of Thirteen Industrial and Developing Countries London and Boston Butterworths

Leach M [1985] Responding to Fuelwood Scarcity Outside Intervention and the Household View in Two Maasai Villages in Tanzania PhD dissertation University of Cambridge England

Leckie J ed [1915] Other Homes and Garbage Designs for SelfshySufficient Living San Fransico Sierra Club Books

Madon G [1982] Principaux resultats de Itevaluation du taux dtutilisation des cuisinieres ameliorees tBan ak Sauft construites en zone rurale tI CERER Dakar Senegal Cited in Cecelski [1984]

Mahmud I and HN Islam (1982) Energy and End-Use Issues in Bangladesh Center for Energy and Environmental Studies Princeton University Princeton New Jersey

- 180 -

Manibog Fernado [1984] ItImproved Cooking Stoves in Developing Countries Problems and Opportunities Energy 9 197-227

McSweeney B [1980] ItCollection and Analysis of Data on Rural Womens Time Use Studies in Family Planning 10 11-12 Population Council New York

Morgan WB [1983] Urban Demand Studying the Commercial Organisation of Wood Fuel Supplies In Ref FAO [1983c]

Munasinghe M [1983] Semi-annual Progress Report Office of the Senior Energy Adviser to the Director of Power amp Energy Colombo Sri Lanka

NAS [1980] Firewood Crops Shrub and Tree Species for Energy Production Washington DC National Academy of Sciences

NCAER [1959] Domestic Fuels in India New Delhi India National Council of Applied Economic Research

Natarajan I [1985] Domestic Fuel Survey with Special Reference to Kerosene (2 vols) New Delhi India National Council of Applied Economic Research

[1986] Personal Communication - National Council of Applied Economic Research New Delhi

Neelakantan J [1975] Assessment of Availability of Cow Dung for Potential Gobar Gas Production in India Agricultural Situation in India

Newcombe K [1980] Energy for Development the Energy Policy Papers of the LAE Project Report 4 UNESCoUNDP Papua New Guinea Human Ecology Program Center for Resource and Environmental Studies Australian National University Australia

[1984a] ItFirewood Supply and Demand in a Papua New Guinea Highlands Village Energy Department The World Bank Washington DC Processed (Restricted)

[1984b] An Economic Justification for Rural Afforestation The Case of Ethiopia Energy Department Paper No 16 The World Bank Washington DC

[1985] The Commercial Potential of Agricultural Residue Fuels Case Studies on Cereals Coffee Cotton and Coconut Crops Energy Department Paper No 26 The World Bank Washington DC

- 181 shy

lIkonoki SR [1983] The Energy Crisis of the Poor in Tanzania Institute of Development Studies University of Dar es Salaam Tanzania

Openshaw K [1978] Woodfuel a Time for Reassessment Natural Resources Forum 3(1)

[1983] Measuring Fuelwood and Charcoal In FAa [1983c)

Osse L [1974) Charcoal Wood [Firewood] Charcoal and Carbonization UNDPFAO Report DPARG70536

OTA [1980] Energy from Biological Processes Vol 2 Technical and Environmental Aspects Washington DC Office of Technology Assessments

Parikh J [1982] Rural Household Energy Consumption in Bangladesh a Critical Assessment Paper for EEC course on Women and the International Development Strategy Vienna Austria

Pathak BS and AK Jain [1984] Characteristics of Crop Residues Report on the Physical Chemical and Thermochemical Properties of Ten Selected Crop Residues Punjab Agricultural University Ludhiana India

PFI [1981] Production of Energy with 10 Million Rupees Peshawar Pakistan Forest Institute (Internal memorandum)

PME [1982] Energy Sector Survey Series 1979 Urban Household Energy Demand Manila Philippines Ministry of Energy

Prasad K Krishna [1982] Woodburning Stoves Geneva International Labor Office Technical and Employment Branch

Prasad K Krishna and Ernst Sangren [1983) Technical Aspects of Woodburning Cookstoves Woodburning Stove Group Eindhoven University of Technology The Netherlands

Prasad K Krishna E Sangren M Sielcken and P Visser [1985J Test Results on Kerosene and Other Stoves Energy Department Paper No 27 The World Bank Washington DC

Prasad K Krishna and P Verhaart eds [1983J Wood Heat for Cooking Bangalore India Indian Academy of Sciences

Quader AK and KI Omar [1982J Resources and Energy Potentials in Rural Bangladesh - a Case Study of Four Villages Bangladesh University of Engineering amp Technology Dacca Commonwealth Science Council London England

d

- 182 -

Reddy AKN et a1 [1980] Rural Energy Consumption Patterns a Field Survey Indian Institute of Science Bangalore India

Samantha [1982] Social Forestry in Orissa Case Study on Fuel Availability and its Impact on Womens Time Disposition and Lifestyle Operations Research Group Bhubaneswar India

SAR [1980] Nepal Community Forestry Development and Training Project South Asia Projects Department The World Bank Washington DC (Restricted)

[1983] India Karnataka Social Forestry Project The World Bank South Asia Projects Department Washington DC (Restricted)

Sathaye J and S Meyers [1985] Energy Use in Cities in Developing Countries Lawrence Berkeley Laboratory Berkeley Processed

Saunier GY et al [1983] Evaluation and Selection of LignoshyCellulose Wastes Which Can be Converted into Substitute Fuels Bangkok Asian Institute of Technology

Schipper Lo et a1 [1982] International Residential Energy Endshyuse Data Analysis of Historical and Present Day Structure and Dynamics Energy 7 5-21

Schramm G and D Jirhad 1984] Sub-Saharan Africa Policy Paper Energy Energy Department The World Bank Washington DC (Restricted) bull

Sepp C et ale [1983] Un foyer metallique a un trou pour la Haute-Volta Informations No5 (avril-juin 1983) Association Bois de Feu Marseilles France

SERU [1981] The Use of Energy and Attitudes Towards Energy Conservation Among Urban Households Kuala Lumpur Malaysia Socio-Economic Research Unit and Prime Ministers Office

SFMAB [1982] Social Forestry Project in Tamil Nadu Report on Preliminary Surveys (Second version) Social Forestry Monitoring Advisory Board Madras India

Shiva v et a1 [1981]= Social Economic and Ecological Impact of Social Forestry in Kolar Indian Institute of Management Bangalore India

Siwatibau Suliansa [1981] Rural Energy in Fiji a Survey of Domestic Rural Energy Use and Potentia1 International Development Research Center Ottawa Canada

- 183 -

Skaar C [1972] Water in Wood Syracuse University Press Syracuse NY

Skar SL [1982] Fuel Availability Nutrition and Womens Work in Highland Peru Rural Employment Policy Research Program International Labor Organisation Geneva Switzerland

Skutsch M [19841 Why People Dont Plant Trees The Socioeconomic Impacts of Existing Woodfuel Programs -- Village Case Studies Tanzania Washington DC Resources for the Future

Smith KR (1985] Biomass Fuels Air Pollution and Health Ambio 14[451 285

Smith KR JM Last and HW De Koning [1984] Biomass Fuel Combustion and Health Bulletin of the World Health Organization 63(1)11-26

Spears J [1978] Wood as an Energy Source the Situation in the Developing World Paper to 103rd annual meeting of American Forestry Association Washington DC

Strasfogel S [1983a] Programme regional foyers ameliores repubUques des bles du Cap-Vert Comite Permanent Interetats de Lutte Contre la Secheresse dans la Sahel Ougadougou Burkina Faso

[1983b] Compte rendu resultats recommandations Mission dappui et de suivi technique foyers ameliores au Niger CILSS Niger

Strout AM [1978] The Demand for Kerosene in Indonesia Unpub report cited in Directorate General for Power [1981) Energy Planning for Development in Indonesia Jakarta Indonesia

Stuckey D [1983] Technology Assessment Study of Biogas in Developing Countries Duebendorf Switerzland International Reference Centre for Waste Disposal

Sulitatu w c Kris-Spit and P Bussmann [1983] The Tamil Nadu Metal Stove WSG ThetApeldoorn The Netherlands

Tandler J and M Kendis [19841 Biomass Densification Energy Department The World Bank Washington DC (Restricted)

Timberlake L [19851 Africa in Crisis The Causes the Cure of Environmental Bankruptcy London Earthscan International Institute for Environment and Development

sa

- 184 shy

UNDPThe World Bank [1982a] Burundi Issues and Options in the Energy Sector Energy Department Report No 3778-BU Washington DC (Restricted)

[1982b] Kenya Issues and Options in the Energy Sector Energy Department Report No 3800-KE Washington DC (Restricted)

[1982c] Malawi Issues and Options in the Energy Sector Energy Department Report No 3903-MAI Washington DC (Restricted)

[1983a] The Gambia Issues and Options in the Energy Sector Energy Department Report No 4743-GM Washington DC (Restricted)

[1983b] Nepal Issues and Options in the Energy Sector Energy Department Report No 4474-NEP Washington DC (Restricted)

[1983c] Nigeria Issues and Options in the Energy Sector Energy Department Report No 4440-UNI Washington DC

[1984a] Costa Rica Issues and Options in the Energy Sector Energy Department Report No 4655-CR Washington DC (Restricted)

[1984b] Ethiopia Issues and Options in the Energy Sector Energy Department Report No 4741-ET Washington DC (Restricted)

[1984c] Liberia Issues and Options in the Energy Sector Energy Department Report No 5279-LBR Washington DC (Restricted)

[1984d] Morocco Issues and Options in the Energy Sector Energy Department Report No 4157-MOR Washington DC (Restricted)

[1984e] Peru Issues and Options in the Energy Sector Energy Oepartment Report No 4677-PE Washington DC (Restricted)

____~ [1984f] Niger Issues and Options in the Energy Sector Energy Department Report No 4642-NIR Washington DC (Restricted)

[1984g] Yemen Arab Republic Issues and Options in the Energy Sector Energy Department Report No 4157-MOR Washington DC (Restricted)

t

- 185 shy

[1984h] Energy Issues and Options in Thirty Developing Countries Energy Department Report No 5230 Washington DC (Restricted)

[1985a] Madagascar Issues and Options in the Energy Sector Energy Department (Restricted)

[1985b] Thailand Rural Energy Issues and Options Report No 5334-Th Energy Department Washington DC (Restricted) bull

[1986] Syria Issues and Options in the Energy Sector Energy Department Report No 5822-SYR Washington DC (Restricted)

[1987] Niger Improved Wood Stoves Project Energy Department Washington DC Processed (Restricted)

Van Buren A [1984] Wood Fuel Commerce in Nicaragua Final Report of SIDAIIED Woodfuel Commercialization Project London International Institute for Environment and Development

Vennetier P [1980] La consommation de lenergie traditionelle en milieu Africain lexemple de Ngaoudere Cameroun In Anon 1981

VITA [1984] Testing the Efficiency of Wood-Burning Cook Stoves Provisional International Standards Volunteers in Technical Assistance Arlington Virginia

Wade Herbert A [1983] The Use of Photovoltaic Systems for Rural Lighting An Economic Analysis of the Alternatives Proceedings of the International Solar Energy Society Solar World Congress Perth Australia

Weatherly P [1980] Environmental Assessment of the Rural Electrification I Project (Indonesia) US Agency for International Development Washington DC

Weber F [1982] Review of CILSS Forestry Sector Program USAID Washington DC

White B [1976] Population Involution and Employment in Rural Java Development and Change 7275

Wijesinghe LCA [1984] A Sample Study of Biomass Fuel Consumption in Sri Lanka Households Biomass 5261-82

- 186 shy

[1984] A Quantitative Study of the Direct Use of Kerosene for Lighting in Sri Lanka Households Natural Resources Science Authority of Sri Lanka Colombo Processed

Wood TS [1981] The Hazards of Testing Improved Wood-burning Cookstoves tI NAS Bostid Development 1(4)4-06

World Bank 1980] Renewable Energy Resources in the Developing Countries Washington DC

World Bank [1983] Energy Transition in Developing Countries Washington DC

WRI [1985] Tropical Forests A Call for Action Washington DC

Cover13
Abstract
Contents13

under the guidance of Richard Dosik Rene Moreno

Willem Floor Mikael Grut Fernando Manibog and Kenneth Newcombe

The World Bank Washington DC

- iii shy

ABSTRACT

- iv shy

- v shy

TABLE or CORTEIITS

-~~~~ ~~---------------------~-------

vi

CHAPTER III

CHAPTER IV

61 61 61 61 65 65 67 67 69 70 72 73 74 74 80 82 83

85 85 86 86 87 88 88 92 92 93

93

95 97 98 98

101 102 104 107 107 109 111 112 114 117

117 120 122 122 124

- vii shy

E Case Studies 143

- viii shy

- ix shy

INTRODUCTION

- 2 shy

- 3 shy

- 4 shy

- 5 shy

- 6 shy

CHAPTER I

- 1 shy

B Primary Energy ~

C Secondary Energy

Kerosene

Production well

Refinery

(Distribution amp

Marketing)

Estimates uncertain

Energy use losses

- 8 shy

- 9 shy

Measurement Units

- 10 shy

1 bull Heat NHV

Output 1

- 11 shy

W= D(l + D100) D = W(l - W100)

- 12 shy

Fiber

Ash

Water

-

~

Net D

High E

DryG

Wet BWater

A losses

F Water

World Bank-307366

HEAT VALUE FORMULAE

- 13 shy

25

30

~ ~ 20 11

2 a

15 ic 0

15 ~ ~ ~ 8 u u i

10 J

~ ~ middot0

o 20 40 60

5

- 14 shy

- 15 shy

Heating Value

(MJkg)

20

GHV

NHV

I I MCWBo

I 30 40 60 80 100 MCDB

World Bank-307368

10 20

- 16 shy

= Vo~ume (m )

and since

then

- 11 shy

- 18 shy

Water loss 4658 kg

Water loss 13333 kg

________________________~-----~~~~------~w~a~~-r----~ ~-----66-6-7-k-9----~ 200 kg

I-

World Bank-307369

(point of use)

10 111

12000 11060 (1659)

ENERGY STAGE

Content (lIcdb)

(NHV MJkg)

Basic Data

Harvest Primary

12658 12658

526 111

12000 9620

(750)

Basic density

approx 1 800

approx 800

25 333

12000 10744 (1343)

600 kgm3

- 19 shy

-~~----------------------

- 20 shy

= Stove Cooker

030 (generation)

013

10

030

10

060

10

015 026 014

- 21 shy

Contents (E)

World Bonk-30736 10

- 22 shy

- 23 shy

- 24 shy

D BASIC STATISTICS

Data Validity

187 145 090 160 167

150 5

0367

069 197 091 068 053 141 088 085 099

8 0475

- 25 shy

Elasticities

as it of

FA b d1 d2 d3 d7

- 26 shy

- 27 -

FIGURE 16

500 -----------------------------------------------

400

300

200

100

fIIII-- fIIIIfIIII

~ -shy

--------shy-

LPG ~ ~ ~ ~

~ ~

~

o ~__________________________________________~

1974

1984

08 r-----------------------------

07 06

Kerosene Price - I

05 - - I - I shy

04

03

02 ~______________________~

1974 1984

World Bonk-31074

- 28 shy

CHAPTER II

- 29 shy

B DATA RESOURCES

- 30 shy

- 31 shy

Local Micro Surveys

- 32 shy

(kgyear)

373 232

246 161

77 87

0 3

0 0

- 33 shy

Supply variables

- 34 shy

Demand variables

o Household income

o Household size

005- 01 up to 15

05 - 1 up to 15

- 35 shy

- 36 shy

Countrysurvey

Togo LOIIe (income)

Average

35

24 33 59

30

11 2

83 236

30

51

Rllnge GJ

34 - 35

21 - 29 31 - 39 57 - 66

27 - 48

135 - 337

23 - 38

46 - 55

bull Biomass

49 - 67

26 - 72 36 - 78 70 - 84

25 - 80

79

41 lt1

22 - 87

Source

Parikh [19821

FBS (1983)

Newcombe [1980)

DeS (1983)

Grut [19711

- 37 shy

- 38 shy

(1 vi II age) 22 94 - 38 Spears [1978)

- 39 shy

- 40 shy

- 41 shy

bullbull bull

bull bull bull

- 42 shy

Rural so

Srazil

~

J bull bull

bull I

I bull

bull

~ I (

I

OL--L~__L--L~__~~~__~~~__~~~__~~~~

bull

bull _ Pakistan

India

bull -- - - ~r- _~ ~ ~ ------------------------------~B~ra~zil

bullbull

- 43 shy

Rurol 100

Indio

Pakistan CD

805s () gtshy ~ w

QZ J

~ in

o 2 4 6 8 10 12 14 16 18

Urban

80

I India

20

o 2 4 6 8 10 12 14 16 18

World Bonk-307362

- 44 shy

rural areas but

- 45 shy

biofuels -

other

to saturate at the

increases shortages

Household Size

- 46 shy

-----

---------

- 47 shy

f

Total

1

_ Per Capita

Household Size --

f ~ ltJ)c w

Total

Per Capita

Household Size ---t

World Bank-307363

bull bullbull

- 48 shy

Rural

bull 10 -

Sri Lanka bull8

( Q)

~ (] gt 6 Indio

~ c bull

1bull~ -_ shyw _-shy __ ~ 0 0 4 U (j) 0

2

o 2 4 6 8 10 12 14 16 18

Urban

gt 60gt ee

(j) c w

--r ----shy~ ---__-_ - 2

O~~~__~~~__L-~~__L-~~__L-~~__L-~~~

o 2 4 6 8 10 12 14 16 18

World Bank-3073611

- 49 shy

and 0 = e Nf

- 50 shy

UK 1982 62 119 43 DOE ( 1983)

Egypt 1975 36 46 30 ILO ( 1979)

Lesotho 1973 48 88 37 ILO [ 1979)

Pakistan 1979 40 86 18 FBS [ 1983)

Sri Lanka 1981 47 97 32 DCS [19831

- 51 shy

Energy Prices

in middle income

- 52 shy

15 GJ

- 53 shy

- 54 shy

fuel can emphasize

- 55 shy

- 56 shy

OYer - 500 294 754 135 289 156 173 200 - 500 275 662 94 286 130 142 100 - 200 269 575 92 198 72 213 50 - 100 266 562 80 187 64 225 20 - 50 234 376 63 95 29 188

Under 20 244 390 67 166 1 5 143

All 266 570 93 212 85 177

~ NataraJan [19851

- 57 shy

Firewood 1979 600 409 251 17 4 121 424 1984 535 308 179 99 96 274

Soft Coke 1979 128 202 236 167 17 3 184 1984 64 180 179 152 83 153

Kerosene 1979 132 213 215 220 189 187 1984 238 369 402 382 328 357

LPG 1979 08 46 142 269 329 66 1984 152 97 83 88 101 101

Other 1979 133 131 156 170 188 139 1984 152 97 83 88 101 101

E ENERGY END-USES

- 58 shy

Cooking 826 585 503 854 797 576 833 826 489

Water heating 20 91 340 105 367 43 422

Space heating 653 324 157 91 98 57 70 131 89

- 59 shy

- 60 shy

F SUMMARY

- 61 shy

CHAPTER III

B COOKING

Consumption Ranges

- 62 shy

(par-boiling rice)

(114) Bialy 119791

225 Evans [19841

- 63 shy

- 64 shy

Charcoal stove 133 140 131 151 199

primus 69 71 10 50 45 wick 21 41 30 48 36

primus 58 57 70 57 130 wick 59 55 63 60 147

- 65 shy

Fuel Preferences

- 66 shy

- 67 shy

Stove Types

- 68 shy

- 69 shy

=c Value

Kerosene Wick

Pres sur i zed ( 0U ) 23 - 65 25 - 55 40

Gas (LPG) Butane 38 - 65 40 - 60 45

- 70 -

Stove Costs

- 71 shy

Woodburnlng Stoves

Mud 10

Clay 15 20

Metal 060 - 600

Charcoal 10 25

Kerosene 2 - 8

Gas 120

Electric 140

- 72 shy

Rubber Wood

Rice husk

Sawdust

Charcoal

Corn cob

Rice husk log

Sawdust log

Bucket

Rangsit

2-hole mud

l-ga I can

Bucket

----------------------baht-------------------shy

24 16 114 16 130

23 16 119 16 135

16 19 204 30 234

12 19 261 22 266

16 76 576 03 564

18 1 70 646 16 662

14 170 884 16 900

21 145 893 16 909

17 145 1124 16 1140

25 185 1267 16 1283

18 203 1892 16 1908

- 13 shy

D LIGHTING

- 74 shy

- 75 shy

(liters per year)

Rural

Sri Lanka

Thai land

Indonesia 1976

28 143

35 42 52 45-61 25 51

70-500 254 148

34

104 96-140

55-91

31 86

570

FBS [1983

WiJeslnghe (1984)

Natarajan (19851

Strout (19781

- 76 shy

~ lt3 3- 6 6-12

12-18 18 All

Urban lt3 3-6 6-12

12-18 18 All

25 29 41 46 51 28

29 31 31 50 86 31

087 102 144 160 179 097

103 107 107 174 302 108

90 84

104 101 106 91

45 61 48 39 39 53

156 163 205 283 322 178

164 189 243 324 425 217

088 010 088 013 110 015 137 013 153 013 096 011

075 015 089 013 104 011 130 014 167 019 096 012

~ NataraJan [19851

- 77 shy

- 78 shy

(USS)

45 43 29

large It

550 750

1175

- 19 shy

Incandescent

40 W bulb 480 60 Wbulb 850

100 W bulb 1500

Fluorescent tubes

55 W bulb 2250

120 143 149

1000 1000 1000

357 556

5000 5000

476 385

625

5000 6000 7500

41 ~7 5000

life of 20000 hours

05 05 06

130 al 130 al

130 92

250

120

- 80 shy

TOTAL

480 900 21 11

1000 5000 05 130 a

102 11 51 143

83 18 415 90

51 143 ~ 27

17 55 17 0

51 143 2075 45

2585 188

- 81 shy

Example

- 82 shy

- 83 shy

F SPACE HEATING

I shy

- 84 shy

Average Delivered

Energy for

Time c Period

B

World Bonk-31214

- 85 shy

CHAPTER IV

- 86 shy

- 87 shy

Access to Resources

- 88 shy

- 89 shy

- 90 shy

- 91 shy

Benefit Type

hydropower systems

- 92 shy

- 93 shy

- 94 shy

Supply Factors

Clear Fell ing

- 95 shy

- 96 shy

Year ($) ($) (m3) (Im3) ($) ($)

1 330 - 330 2 165 - 165 3 130 - 130 4-5 60 - 60 6 60 37 209 353 738 + 641 7-10 60 - 60

11 60 81 456 530 2417 +2276 12-15 60 - 60 16 60 73 343 706 2422 +2289 17-19 60 - 60 20 60 375 1515 882 13362 +12927

TOTALS 1645 566 2523 18939 +16728

General data

~ PFI (1981)

- 97 -

(yrs) (m3hayr)

1st year 2nd year

E citriodesra ~I

10 - 12 4 - 8

7 - 10

8 - 10 6 ms 2 - 5

5 - 15 5 - 10 5 - 10

10

4 - 5

25 - 30 15 - 20 10 - 15 8 5 - 7

7 - 10 14 - 15 8

10 15

17 - 20 10 - 15

5 - 20 35 - 60

10 - 20

25 - 60 15 odthayr 20 - 25

10 - 30 40

17 - 45 5 - 7

15 - 10

05 - 10 5 - 10 5

10

10 - 15

20 - 30 2 - 11

15

7 - 10 5 - 6

06 - 08 06 - 08

05 - 08 05 - 08

08 - 12

03 04

08 - 10 04 - 05 04 - 05 04 - 05

(lIght)

(heavy) (heavy) 05 - 060 06 - 09 06 - 08

06 06 08 - 11

07-10 07-10

Source NAS [19801

- 98 -

- 99 shy

- 100 shy

Charcoal 34 09 24 14 16

- 101 shy

Fuel

(I)

Del ivered Price

(kunit)

(2)

Net HV (MJunit)

()

(4) Effective Price

(kMJ uti I ized)

Appl lance Cost

(N=IOOk)

LPG Electricity

1471kg 251kg 3481 3481 490kg 36kWh

8-13 20-25 30-40 30-40 45-55 60-70

89 -44 -07 -02 -13 -24 -

145 58 10 27 15 27

na na 3 al

38 bl 40 45 40

(Col 2) x (Col 3)100

- 102 shy

Plantation Costs

=

1984 US$ I

0 bamboo w

- 104 shy

- 105 shy

- 106 shy

- 107 shy

E CHARCOAL

- 108 shy

Charcoal yields

- 109 shy

- 110 shy

Thin Bark

Compact Heavy

Correct Dimensions

Healthy

Low Mol sture

Source Osse (1974)

- 111 shy

Asia Thai land (1984) 009 - 021 29 03 - 07 23 13 - 30

- 112 shy

husk 16 - 45

- 113 shy

Alfalfa straw

Groundnut shells

Maize stalks

Maize cobs

Rice husks

(1 )

(1)

(2) (3)

(3)

(1) (4) ( 1 )

(4) (1)

(4) ( 1 )

(4)

(5) (4)

(5) (4) (1)

(2) (1)

(I)

(1)

( 4)

()

48

08 60

172 33

44 64 34 15 18 32 20J

192

165 149

11

85

(MJkg)

184 173 194 183 201 181 158 174 197 200 182 167 189 17 4

214 186 152 150 153 155 168 194 210 211 189 17 2

- 114 shy

- 115 shy

- 116 shy

Rice - 100 + 800 + 430

Wheat o + 600 + 290

Maize + 100 +1300 + 480

Millet o + 500 + 250

~ ICAR (1984)

- 117 shy

- 118 shy

FeedstocKs Reason

Good

Poor

high I ign in

(KWht)

Product Density

(tem3)

(tehour)

Range of Systell

Costs (US$OOOte h)

(US$ OOOte h)

30-60 NA NA

015-08 100 - 15

20shy 60 25 - 110

40 30

- 75 - 40

--

- p I I low-shaped

20 - 35 NA 20 - 60 130 - 300 30 - 60 ----

()

I

- 120 shy

- 121 shy

Residue (1) (2) (3)

Grinding

723 (422) (135 ) (150) (016)

1180 (556) (1 76) (437) (011)

10 2005 2502

1941

(338) (1403) (201)

4443

173 ( 12)

2257

1903 (1040) (411) (432) (020)

144

854 (237) (52S) (080) (012)

088 2989 3215

1941

(338) (1403) (201)

5156

150 (15)

344

1085 (239) ( 1 64) (640) (042)

144

8S4 (237) (S2S) (080) (012)

088 2171 2735

1941

(338) ( 1403) (201)

4676

174 (15)

269

- 122 shy

G ANIMAL WASTES

Direct Combustion

Animal

(kgyr)

(kg)

Cattle 27000 200 5400 80 1000 24 12

Pigs 30000 50 1500 80 300 315 18

Poultry 9000 15 13 60 5 63 63

- 124 shy

Biogas

- 125 shy

- 126 shy

CHAPTER V

B DATA SOURCES

- 127 shy

- 128 shy

cycle costs

- 129 shy

Supply Data

- 130 shy

- 131 shy

land use

As above

- 132 shy

- 133 shy

1980 1985 1990 1995 2000 2005

Standing stock 000 3 17 500 16010 13837 10827 6794 1520

Consumption 000 3yr 600 696 806 935 1084 1256

Deficit 000 3yr 250 376 529 718 948 1226

(Population ooos) (1000) ( 1 159) (1344) (1558) (1806) (2094)

Assumptions

Calculation method

- 134 shy

- 135 shy

1980 1985 1990 1995 2000 2005

open fire 75 663 504 33 196 10 stove 5 117 216 33 364 35

- 136 shy

1980 1985 1990 1995 2000 2005

- 137 shy

- 138 shy

1980 1985 1990 1995 2000 2005

500 17500

580 15907

672 14061

779 11920

903 9439

1047 6562

TOTAL 647 663 680 701 725 753

- 139 shy

Wood Consumption

~~5~ 43 2

Annual Increase 0

0 O~________L-________~________~________-L________~

1980 1985 1990 1995 2000 2005

100r--__bullbull

~~====3--- 2

1

0

o~--------~--------~----------~--------~--------~ 1980 1985 1990 1995 2000 2005

World Bank-307364

- 140 shy

- 141 shy

- 142 shy

Land type 2 3 4 5 6

Land type

Altitude (m)

200-1000 0- 500

500-1500 1000-2000

1000-2500 1500-3000

Rainfall (mm)

lt400 500-1000 500- 900

800-1200

1 000-1 300 lt1200

- 143 shy

E CASE STUDIES

Average Both fuels

548

2344 1148 3491

1482 362

1844

3826 1510 5336

- 144 shy

( of far Itany)

1145 15043 I 2850 21274 28137 44620

609 55

77 6 67

1021 119

29 34 13 14 41 27

Tota I 123069 2648 ~ 21

TOTAL 6750 3934 1560 5494 (1256)

- 145 shy

198] 1985 1987 1989 1991 993 995

Sub-Total A 00 00 530 797 13 2228 3A46 to J

- 147 shy

Annex 1

( mcwb) (MJkg)

Biomass Fuels

Fossil-Fuels

lignite Peat

Coke

Petroleum coke

40 20 15 0 5

50 13 12 9

12 12 11 24 12 40 13 12

5 5 5

10-9

155 66

200 290 82

162 160 144 152 147 154 119 164 98

179 120

31~4

293 188

113 146

201 239 218

285

352

- 148 shy

(MJkg) (t-tJ1 itre)

Fossil Fuels

Crude 01 I 086 419 367

gas

- 149 shy

(MJm3)

Refinery Gas 461

- 150 shy

Annex 2

SI American

II EnerSI Symbols

SI

AmericanGeneral

- 151 shy

W v V a A

kVA

BTUhr hp

bd bId bdoe

g kg lb lbs

tpa tpy

m km mi

2sq m mha ac

1 3cu m m

gal

SCF CF

V Biomass amp Other

od 00 odt ODT

ad AD mcwb mcdb

MAl GHV NHV

SI

AmericanGeneral

2

cubic metre 1 10000 -3 28317 -2 76455 -1 28413 -5 56826 -4 45461 -3 29574 -5 47318 -4 37854 -3 itre 99997 +2 1 28316 +1 76453 +2 28412 -2 56825 -1 45460 0 29573 -2 47316 -1 37853 0

US gallon

21134 26417

+3 +2

21134 26418

0 -1

59844 74805

+1 0

16158 20197

+3 +2

60047 75059

-2 -3

12009 15012

0 -1

960761 12009

0 0

62500 78125

-2 -3

I 12500 -1

80000 0 w

10000 22046 98421 11023

-3 0

-4 -3

10000 1

22046 98421 11023

+3

+3 -1 0

45359 45359

44643 50000

-1 -4

-4 -4

10160 10160 22400

11200

+3 0

+3

0

90718 90718 20000 89286

+2 -1 +3 -1

27778 37251

-7 -7

11630 15596

-3 -3 13410 0

74570 -1 29307 39301

-4 -4

U1 po

per minute

10000 13596

13410 14331

56869

-3 -3

-3 -2

-2

10000

13596

13410 14331

56869

+3

0

0 +1

+1

73550 73550

98632 10540

41827

+2 -1

-1 +1

+1

74570 74570 10139

1 10686

42407

+2 -1 0

+1

69780 69780 94874

93577

39683

+1 -2 -2

-2

0

17584 17584 23908

23581 25200

+1 -2 -2

-2 -1

x 10 -4

- 155 shy

Annex 4

GLOSSARY

Air-dried weight

Anaerobic processes

Bagasse

Biogas

Biomass fuels

Calorie

Coal equivalent

- 156 shy

Coal replacement

- 157 shy

- 158 shy

Utilized energy

Green weight

Annex 5

CLASS Material

Jl 0

Owner-bu i It

tinerant art isan

Industrial

I- 0shyo

marketing approach

t- 0 t-

- 162 shy

Annex 6

- 163 shy

- 164 shy

Annex 7

r I Payback Period

~-~ Time

World Bonk-307365

- 165 shy

I + F x P = f x p

- 166 shy

Annex 8

- 167 shy

- 168 shy

- 169 shy

Annex 9

for Firewood

o

are Breached

World Bank-3073612

- 171 shy

- 172 shy

- 173 shy

BIBLIOGRAPHY

- 174 -

- 115 -

- 176 -

- 177 -

- 178 shy

- 179 -

- 180 -

- 181 shy

d

- 182 -

- 183 -

sa

- 184 shy

t

- 185 shy

- 186 shy

Cover13
Abstract
Contents13

- iii shy

ABSTRACT

- iv shy

- v shy

TABLE or CORTEIITS

-~~~~ ~~---------------------~-------

vi

CHAPTER III

CHAPTER IV

61 61 61 61 65 65 67 67 69 70 72 73 74 74 80 82 83

85 85 86 86 87 88 88 92 92 93

93

95 97 98 98

101 102 104 107 107 109 111 112 114 117

117 120 122 122 124

- vii shy

E Case Studies 143

- viii shy

- ix shy

INTRODUCTION

- 2 shy

- 3 shy

- 4 shy

- 5 shy

- 6 shy

CHAPTER I

- 1 shy

B Primary Energy ~

C Secondary Energy

Kerosene

Production well

Refinery

(Distribution amp

Marketing)

Estimates uncertain

Energy use losses

- 8 shy

- 9 shy

Measurement Units

- 10 shy

1 bull Heat NHV

Output 1

- 11 shy

W= D(l + D100) D = W(l - W100)

- 12 shy

Fiber

Ash

Water

-

~

Net D

High E

DryG

Wet BWater

A losses

F Water

World Bank-307366

HEAT VALUE FORMULAE

- 13 shy

25

30

~ ~ 20 11

2 a

15 ic 0

15 ~ ~ ~ 8 u u i

10 J

~ ~ middot0

o 20 40 60

5

- 14 shy

- 15 shy

Heating Value

(MJkg)

20

GHV

NHV

I I MCWBo

I 30 40 60 80 100 MCDB

World Bank-307368

10 20

- 16 shy

= Vo~ume (m )

and since

then

- 11 shy

- 18 shy

Water loss 4658 kg

Water loss 13333 kg

________________________~-----~~~~------~w~a~~-r----~ ~-----66-6-7-k-9----~ 200 kg

I-

World Bank-307369

(point of use)

10 111

12000 11060 (1659)

ENERGY STAGE

Content (lIcdb)

(NHV MJkg)

Basic Data

Harvest Primary

12658 12658

526 111

12000 9620

(750)

Basic density

approx 1 800

approx 800

25 333

12000 10744 (1343)

600 kgm3

- 19 shy

-~~----------------------

- 20 shy

= Stove Cooker

030 (generation)

013

10

030

10

060

10

015 026 014

- 21 shy

Contents (E)

World Bonk-30736 10

- 22 shy

- 23 shy

- 24 shy

D BASIC STATISTICS

Data Validity

187 145 090 160 167

150 5

0367

069 197 091 068 053 141 088 085 099

8 0475

- 25 shy

Elasticities

as it of

FA b d1 d2 d3 d7

- 26 shy

- 27 -

FIGURE 16

500 -----------------------------------------------

400

300

200

100

fIIII-- fIIIIfIIII

~ -shy

--------shy-

LPG ~ ~ ~ ~

~ ~

~

o ~__________________________________________~

1974

1984

08 r-----------------------------

07 06

Kerosene Price - I

05 - - I - I shy

04

03

02 ~______________________~

1974 1984

World Bonk-31074

- 28 shy

CHAPTER II

- 29 shy

B DATA RESOURCES

- 30 shy

- 31 shy

Local Micro Surveys

- 32 shy

(kgyear)

373 232

246 161

77 87

0 3

0 0

- 33 shy

Supply variables

- 34 shy

Demand variables

o Household income

o Household size

005- 01 up to 15

05 - 1 up to 15

- 35 shy

- 36 shy

Countrysurvey

Togo LOIIe (income)

Average

35

24 33 59

30

11 2

83 236

30

51

Rllnge GJ

34 - 35

21 - 29 31 - 39 57 - 66

27 - 48

135 - 337

23 - 38

46 - 55

bull Biomass

49 - 67

26 - 72 36 - 78 70 - 84

25 - 80

79

41 lt1

22 - 87

Source

Parikh [19821

FBS (1983)

Newcombe [1980)

DeS (1983)

Grut [19711

- 37 shy

- 38 shy

(1 vi II age) 22 94 - 38 Spears [1978)

- 39 shy

- 40 shy

- 41 shy

bullbull bull

bull bull bull

- 42 shy

Rural so

Srazil

~

J bull bull

bull I

I bull

bull

~ I (

I

OL--L~__L--L~__~~~__~~~__~~~__~~~~

bull

bull _ Pakistan

India

bull -- - - ~r- _~ ~ ~ ------------------------------~B~ra~zil

bullbull

- 43 shy

Rurol 100

Indio

Pakistan CD

805s () gtshy ~ w

QZ J

~ in

o 2 4 6 8 10 12 14 16 18

Urban

80

I India

20

o 2 4 6 8 10 12 14 16 18

World Bonk-307362

- 44 shy

rural areas but

- 45 shy

biofuels -

other

to saturate at the

increases shortages

Household Size

- 46 shy

-----

---------

- 47 shy

f

Total

1

_ Per Capita

Household Size --

f ~ ltJ)c w

Total

Per Capita

Household Size ---t

World Bank-307363

bull bullbull

- 48 shy

Rural

bull 10 -

Sri Lanka bull8

( Q)

~ (] gt 6 Indio

~ c bull

1bull~ -_ shyw _-shy __ ~ 0 0 4 U (j) 0

2

o 2 4 6 8 10 12 14 16 18

Urban

gt 60gt ee

(j) c w

--r ----shy~ ---__-_ - 2

O~~~__~~~__L-~~__L-~~__L-~~__L-~~~

o 2 4 6 8 10 12 14 16 18

World Bank-3073611

- 49 shy

and 0 = e Nf

- 50 shy

UK 1982 62 119 43 DOE ( 1983)

Egypt 1975 36 46 30 ILO ( 1979)

Lesotho 1973 48 88 37 ILO [ 1979)

Pakistan 1979 40 86 18 FBS [ 1983)

Sri Lanka 1981 47 97 32 DCS [19831

- 51 shy

Energy Prices

in middle income

- 52 shy

15 GJ

- 53 shy

- 54 shy

fuel can emphasize

- 55 shy

- 56 shy

OYer - 500 294 754 135 289 156 173 200 - 500 275 662 94 286 130 142 100 - 200 269 575 92 198 72 213 50 - 100 266 562 80 187 64 225 20 - 50 234 376 63 95 29 188

Under 20 244 390 67 166 1 5 143

All 266 570 93 212 85 177

~ NataraJan [19851

- 57 shy

Firewood 1979 600 409 251 17 4 121 424 1984 535 308 179 99 96 274

Soft Coke 1979 128 202 236 167 17 3 184 1984 64 180 179 152 83 153

Kerosene 1979 132 213 215 220 189 187 1984 238 369 402 382 328 357

LPG 1979 08 46 142 269 329 66 1984 152 97 83 88 101 101

Other 1979 133 131 156 170 188 139 1984 152 97 83 88 101 101

E ENERGY END-USES

- 58 shy

Cooking 826 585 503 854 797 576 833 826 489

Water heating 20 91 340 105 367 43 422

Space heating 653 324 157 91 98 57 70 131 89

- 59 shy

- 60 shy

F SUMMARY

- 61 shy

CHAPTER III

B COOKING

Consumption Ranges

- 62 shy

(par-boiling rice)

(114) Bialy 119791

225 Evans [19841

- 63 shy

- 64 shy

Charcoal stove 133 140 131 151 199

primus 69 71 10 50 45 wick 21 41 30 48 36

primus 58 57 70 57 130 wick 59 55 63 60 147

- 65 shy

Fuel Preferences

- 66 shy

- 67 shy

Stove Types

- 68 shy

- 69 shy

=c Value

Kerosene Wick

Pres sur i zed ( 0U ) 23 - 65 25 - 55 40

Gas (LPG) Butane 38 - 65 40 - 60 45

- 70 -

Stove Costs

- 71 shy

Woodburnlng Stoves

Mud 10

Clay 15 20

Metal 060 - 600

Charcoal 10 25

Kerosene 2 - 8

Gas 120

Electric 140

- 72 shy

Rubber Wood

Rice husk

Sawdust

Charcoal

Corn cob

Rice husk log

Sawdust log

Bucket

Rangsit

2-hole mud

l-ga I can

Bucket

----------------------baht-------------------shy

24 16 114 16 130

23 16 119 16 135

16 19 204 30 234

12 19 261 22 266

16 76 576 03 564

18 1 70 646 16 662

14 170 884 16 900

21 145 893 16 909

17 145 1124 16 1140

25 185 1267 16 1283

18 203 1892 16 1908

- 13 shy

D LIGHTING

- 74 shy

- 75 shy

(liters per year)

Rural

Sri Lanka

Thai land

Indonesia 1976

28 143

35 42 52 45-61 25 51

70-500 254 148

34

104 96-140

55-91

31 86

570

FBS [1983

WiJeslnghe (1984)

Natarajan (19851

Strout (19781

- 76 shy

~ lt3 3- 6 6-12

12-18 18 All

Urban lt3 3-6 6-12

12-18 18 All

25 29 41 46 51 28

29 31 31 50 86 31

087 102 144 160 179 097

103 107 107 174 302 108

90 84

104 101 106 91

45 61 48 39 39 53

156 163 205 283 322 178

164 189 243 324 425 217

088 010 088 013 110 015 137 013 153 013 096 011

075 015 089 013 104 011 130 014 167 019 096 012

~ NataraJan [19851

- 77 shy

- 78 shy

(USS)

45 43 29

large It

550 750

1175

- 19 shy

Incandescent

40 W bulb 480 60 Wbulb 850

100 W bulb 1500

Fluorescent tubes

55 W bulb 2250

120 143 149

1000 1000 1000

357 556

5000 5000

476 385

625

5000 6000 7500

41 ~7 5000

life of 20000 hours

05 05 06

130 al 130 al

130 92

250

120

- 80 shy

TOTAL

480 900 21 11

1000 5000 05 130 a

102 11 51 143

83 18 415 90

51 143 ~ 27

17 55 17 0

51 143 2075 45

2585 188

- 81 shy

Example

- 82 shy

- 83 shy

F SPACE HEATING

I shy

- 84 shy

Average Delivered

Energy for

Time c Period

B

World Bonk-31214

- 85 shy

CHAPTER IV

- 86 shy

- 87 shy

Access to Resources

- 88 shy

- 89 shy

- 90 shy

- 91 shy

Benefit Type

hydropower systems

- 92 shy

- 93 shy

- 94 shy

Supply Factors

Clear Fell ing

- 95 shy

- 96 shy

Year ($) ($) (m3) (Im3) ($) ($)

1 330 - 330 2 165 - 165 3 130 - 130 4-5 60 - 60 6 60 37 209 353 738 + 641 7-10 60 - 60

11 60 81 456 530 2417 +2276 12-15 60 - 60 16 60 73 343 706 2422 +2289 17-19 60 - 60 20 60 375 1515 882 13362 +12927

TOTALS 1645 566 2523 18939 +16728

General data

~ PFI (1981)

- 97 -

(yrs) (m3hayr)

1st year 2nd year

E citriodesra ~I

10 - 12 4 - 8

7 - 10

8 - 10 6 ms 2 - 5

5 - 15 5 - 10 5 - 10

10

4 - 5

25 - 30 15 - 20 10 - 15 8 5 - 7

7 - 10 14 - 15 8

10 15

17 - 20 10 - 15

5 - 20 35 - 60

10 - 20

25 - 60 15 odthayr 20 - 25

10 - 30 40

17 - 45 5 - 7

15 - 10

05 - 10 5 - 10 5

10

10 - 15

20 - 30 2 - 11

15

7 - 10 5 - 6

06 - 08 06 - 08

05 - 08 05 - 08

08 - 12

03 04

08 - 10 04 - 05 04 - 05 04 - 05

(lIght)

(heavy) (heavy) 05 - 060 06 - 09 06 - 08

06 06 08 - 11

07-10 07-10

Source NAS [19801

- 98 -

- 99 shy

- 100 shy

Charcoal 34 09 24 14 16

- 101 shy

Fuel

(I)

Del ivered Price

(kunit)

(2)

Net HV (MJunit)

()

(4) Effective Price

(kMJ uti I ized)

Appl lance Cost

(N=IOOk)

LPG Electricity

1471kg 251kg 3481 3481 490kg 36kWh

8-13 20-25 30-40 30-40 45-55 60-70

89 -44 -07 -02 -13 -24 -

145 58 10 27 15 27

na na 3 al

38 bl 40 45 40

(Col 2) x (Col 3)100

- 102 shy

Plantation Costs

=

1984 US$ I

0 bamboo w

- 104 shy

- 105 shy

- 106 shy

- 107 shy

E CHARCOAL

- 108 shy

Charcoal yields

- 109 shy

- 110 shy

Thin Bark

Compact Heavy

Correct Dimensions

Healthy

Low Mol sture

Source Osse (1974)

- 111 shy

Asia Thai land (1984) 009 - 021 29 03 - 07 23 13 - 30

- 112 shy

husk 16 - 45

- 113 shy

Alfalfa straw

Groundnut shells

Maize stalks

Maize cobs

Rice husks

(1 )

(1)

(2) (3)

(3)

(1) (4) ( 1 )

(4) (1)

(4) ( 1 )

(4)

(5) (4)

(5) (4) (1)

(2) (1)

(I)

(1)

( 4)

()

48

08 60

172 33

44 64 34 15 18 32 20J

192

165 149

11

85

(MJkg)

184 173 194 183 201 181 158 174 197 200 182 167 189 17 4

214 186 152 150 153 155 168 194 210 211 189 17 2

- 114 shy

- 115 shy

- 116 shy

Rice - 100 + 800 + 430

Wheat o + 600 + 290

Maize + 100 +1300 + 480

Millet o + 500 + 250

~ ICAR (1984)

- 117 shy

- 118 shy

FeedstocKs Reason

Good

Poor

high I ign in

(KWht)

Product Density

(tem3)

(tehour)

Range of Systell

Costs (US$OOOte h)

(US$ OOOte h)

30-60 NA NA

015-08 100 - 15

20shy 60 25 - 110

40 30

- 75 - 40

--

- p I I low-shaped

20 - 35 NA 20 - 60 130 - 300 30 - 60 ----

()

I

- 120 shy

- 121 shy

Residue (1) (2) (3)

Grinding

723 (422) (135 ) (150) (016)

1180 (556) (1 76) (437) (011)

10 2005 2502

1941

(338) (1403) (201)

4443

173 ( 12)

2257

1903 (1040) (411) (432) (020)

144

854 (237) (52S) (080) (012)

088 2989 3215

1941

(338) (1403) (201)

5156

150 (15)

344

1085 (239) ( 1 64) (640) (042)

144

8S4 (237) (S2S) (080) (012)

088 2171 2735

1941

(338) ( 1403) (201)

4676

174 (15)

269

- 122 shy

G ANIMAL WASTES

Direct Combustion

Animal

(kgyr)

(kg)

Cattle 27000 200 5400 80 1000 24 12

Pigs 30000 50 1500 80 300 315 18

Poultry 9000 15 13 60 5 63 63

- 124 shy

Biogas

- 125 shy

- 126 shy

CHAPTER V

B DATA SOURCES

- 127 shy

- 128 shy

cycle costs

- 129 shy

Supply Data

- 130 shy

- 131 shy

land use

As above

- 132 shy

- 133 shy

1980 1985 1990 1995 2000 2005

Standing stock 000 3 17 500 16010 13837 10827 6794 1520

Consumption 000 3yr 600 696 806 935 1084 1256

Deficit 000 3yr 250 376 529 718 948 1226

(Population ooos) (1000) ( 1 159) (1344) (1558) (1806) (2094)

Assumptions

Calculation method

- 134 shy

- 135 shy

1980 1985 1990 1995 2000 2005

open fire 75 663 504 33 196 10 stove 5 117 216 33 364 35

- 136 shy

1980 1985 1990 1995 2000 2005

- 137 shy

- 138 shy

1980 1985 1990 1995 2000 2005

500 17500

580 15907

672 14061

779 11920

903 9439

1047 6562

TOTAL 647 663 680 701 725 753

- 139 shy

Wood Consumption

~~5~ 43 2

Annual Increase 0

0 O~________L-________~________~________-L________~

1980 1985 1990 1995 2000 2005

100r--__bullbull

~~====3--- 2

1

0

o~--------~--------~----------~--------~--------~ 1980 1985 1990 1995 2000 2005

World Bank-307364

- 140 shy

- 141 shy

- 142 shy

Land type 2 3 4 5 6

Land type

Altitude (m)

200-1000 0- 500

500-1500 1000-2000

1000-2500 1500-3000

Rainfall (mm)

lt400 500-1000 500- 900

800-1200

1 000-1 300 lt1200

- 143 shy

E CASE STUDIES

Average Both fuels

548

2344 1148 3491

1482 362

1844

3826 1510 5336

- 144 shy

( of far Itany)

1145 15043 I 2850 21274 28137 44620

609 55

77 6 67

1021 119

29 34 13 14 41 27

Tota I 123069 2648 ~ 21

TOTAL 6750 3934 1560 5494 (1256)

- 145 shy

198] 1985 1987 1989 1991 993 995

Sub-Total A 00 00 530 797 13 2228 3A46 to J

- 147 shy

Annex 1

( mcwb) (MJkg)

Biomass Fuels

Fossil-Fuels

lignite Peat

Coke

Petroleum coke

40 20 15 0 5

50 13 12 9

12 12 11 24 12 40 13 12

5 5 5

10-9

155 66

200 290 82

162 160 144 152 147 154 119 164 98

179 120

31~4

293 188

113 146

201 239 218

285

352

- 148 shy

(MJkg) (t-tJ1 itre)

Fossil Fuels

Crude 01 I 086 419 367

gas

- 149 shy

(MJm3)

Refinery Gas 461

- 150 shy

Annex 2

SI American

II EnerSI Symbols

SI

AmericanGeneral

- 151 shy

W v V a A

kVA

BTUhr hp

bd bId bdoe

g kg lb lbs

tpa tpy

m km mi

2sq m mha ac

1 3cu m m

gal

SCF CF

V Biomass amp Other

od 00 odt ODT

ad AD mcwb mcdb

MAl GHV NHV

SI

AmericanGeneral

2

cubic metre 1 10000 -3 28317 -2 76455 -1 28413 -5 56826 -4 45461 -3 29574 -5 47318 -4 37854 -3 itre 99997 +2 1 28316 +1 76453 +2 28412 -2 56825 -1 45460 0 29573 -2 47316 -1 37853 0

US gallon

21134 26417

+3 +2

21134 26418

0 -1

59844 74805

+1 0

16158 20197

+3 +2

60047 75059

-2 -3

12009 15012

0 -1

960761 12009

0 0

62500 78125

-2 -3

I 12500 -1

80000 0 w

10000 22046 98421 11023

-3 0

-4 -3

10000 1

22046 98421 11023

+3

+3 -1 0

45359 45359

44643 50000

-1 -4

-4 -4

10160 10160 22400

11200

+3 0

+3

0

90718 90718 20000 89286

+2 -1 +3 -1

27778 37251

-7 -7

11630 15596

-3 -3 13410 0

74570 -1 29307 39301

-4 -4

U1 po

per minute

10000 13596

13410 14331

56869

-3 -3

-3 -2

-2

10000

13596

13410 14331

56869

+3

0

0 +1

+1

73550 73550

98632 10540

41827

+2 -1

-1 +1

+1

74570 74570 10139

1 10686

42407

+2 -1 0

+1

69780 69780 94874

93577

39683

+1 -2 -2

-2

0

17584 17584 23908

23581 25200

+1 -2 -2

-2 -1

x 10 -4

- 155 shy

Annex 4

GLOSSARY

Air-dried weight

Anaerobic processes

Bagasse

Biogas

Biomass fuels

Calorie

Coal equivalent

- 156 shy

Coal replacement

- 157 shy

- 158 shy

Utilized energy

Green weight

Annex 5

CLASS Material

Jl 0

Owner-bu i It

tinerant art isan

Industrial

I- 0shyo

marketing approach

t- 0 t-

- 162 shy

Annex 6

- 163 shy

- 164 shy

Annex 7

r I Payback Period

~-~ Time

World Bonk-307365

- 165 shy

I + F x P = f x p

- 166 shy

Annex 8

- 167 shy

- 168 shy

- 169 shy

Annex 9

for Firewood

o

are Breached

World Bank-3073612

- 171 shy

- 172 shy

- 173 shy

BIBLIOGRAPHY

- 174 -

- 115 -

- 176 -

- 177 -

- 178 shy

- 179 -

- 180 -

- 181 shy

d

- 182 -

- 183 -

sa

- 184 shy

t

- 185 shy

- 186 shy

Cover13
Abstract
Contents13

- iii shy

ABSTRACT

- iv shy

- v shy

TABLE or CORTEIITS

-~~~~ ~~---------------------~-------

vi

CHAPTER III

CHAPTER IV

61 61 61 61 65 65 67 67 69 70 72 73 74 74 80 82 83

85 85 86 86 87 88 88 92 92 93

93

95 97 98 98

101 102 104 107 107 109 111 112 114 117

117 120 122 122 124

- vii shy

E Case Studies 143

- viii shy

- ix shy

INTRODUCTION

- 2 shy

- 3 shy

- 4 shy

- 5 shy

- 6 shy

CHAPTER I

- 1 shy

B Primary Energy ~

C Secondary Energy

Kerosene

Production well

Refinery

(Distribution amp

Marketing)

Estimates uncertain

Energy use losses

- 8 shy

- 9 shy

Measurement Units

- 10 shy

1 bull Heat NHV

Output 1

- 11 shy

W= D(l + D100) D = W(l - W100)

- 12 shy

Fiber

Ash

Water

-

~

Net D

High E

DryG

Wet BWater

A losses

F Water

World Bank-307366

HEAT VALUE FORMULAE

- 13 shy

25

30

~ ~ 20 11

2 a

15 ic 0

15 ~ ~ ~ 8 u u i

10 J

~ ~ middot0

o 20 40 60

5

- 14 shy

- 15 shy

Heating Value

(MJkg)

20

GHV

NHV

I I MCWBo

I 30 40 60 80 100 MCDB

World Bank-307368

10 20

- 16 shy

= Vo~ume (m )

and since

then

- 11 shy

- 18 shy

Water loss 4658 kg

Water loss 13333 kg

________________________~-----~~~~------~w~a~~-r----~ ~-----66-6-7-k-9----~ 200 kg

I-

World Bank-307369

(point of use)

10 111

12000 11060 (1659)

ENERGY STAGE

Content (lIcdb)

(NHV MJkg)

Basic Data

Harvest Primary

12658 12658

526 111

12000 9620

(750)

Basic density

approx 1 800

approx 800

25 333

12000 10744 (1343)

600 kgm3

- 19 shy

-~~----------------------

- 20 shy

= Stove Cooker

030 (generation)

013

10

030

10

060

10

015 026 014

- 21 shy

Contents (E)

World Bonk-30736 10

- 22 shy

- 23 shy

- 24 shy

D BASIC STATISTICS

Data Validity

187 145 090 160 167

150 5

0367

069 197 091 068 053 141 088 085 099

8 0475

- 25 shy

Elasticities

as it of

FA b d1 d2 d3 d7

- 26 shy

- 27 -

FIGURE 16

500 -----------------------------------------------

400

300

200

100

fIIII-- fIIIIfIIII

~ -shy

--------shy-

LPG ~ ~ ~ ~

~ ~

~

o ~__________________________________________~

1974

1984

08 r-----------------------------

07 06

Kerosene Price - I

05 - - I - I shy

04

03

02 ~______________________~

1974 1984

World Bonk-31074

- 28 shy

CHAPTER II

- 29 shy

B DATA RESOURCES

- 30 shy

- 31 shy

Local Micro Surveys

- 32 shy

(kgyear)

373 232

246 161

77 87

0 3

0 0

- 33 shy

Supply variables

- 34 shy

Demand variables

o Household income

o Household size

005- 01 up to 15

05 - 1 up to 15

- 35 shy

- 36 shy

Countrysurvey

Togo LOIIe (income)

Average

35

24 33 59

30

11 2

83 236

30

51

Rllnge GJ

34 - 35

21 - 29 31 - 39 57 - 66

27 - 48

135 - 337

23 - 38

46 - 55

bull Biomass

49 - 67

26 - 72 36 - 78 70 - 84

25 - 80

79

41 lt1

22 - 87

Source

Parikh [19821

FBS (1983)

Newcombe [1980)

DeS (1983)

Grut [19711

- 37 shy

- 38 shy

(1 vi II age) 22 94 - 38 Spears [1978)

- 39 shy

- 40 shy

- 41 shy

bullbull bull

bull bull bull

- 42 shy

Rural so

Srazil

~

J bull bull

bull I

I bull

bull

~ I (

I

OL--L~__L--L~__~~~__~~~__~~~__~~~~

bull

bull _ Pakistan

India

bull -- - - ~r- _~ ~ ~ ------------------------------~B~ra~zil

bullbull

- 43 shy

Rurol 100

Indio

Pakistan CD

805s () gtshy ~ w

QZ J

~ in

o 2 4 6 8 10 12 14 16 18

Urban

80

I India

20

o 2 4 6 8 10 12 14 16 18

World Bonk-307362

- 44 shy

rural areas but

- 45 shy

biofuels -

other

to saturate at the

increases shortages

Household Size

- 46 shy

-----

---------

- 47 shy

f

Total

1

_ Per Capita

Household Size --

f ~ ltJ)c w

Total

Per Capita

Household Size ---t

World Bank-307363

bull bullbull

- 48 shy

Rural

bull 10 -

Sri Lanka bull8

( Q)

~ (] gt 6 Indio

~ c bull

1bull~ -_ shyw _-shy __ ~ 0 0 4 U (j) 0

2

o 2 4 6 8 10 12 14 16 18

Urban

gt 60gt ee

(j) c w

--r ----shy~ ---__-_ - 2

O~~~__~~~__L-~~__L-~~__L-~~__L-~~~

o 2 4 6 8 10 12 14 16 18

World Bank-3073611

- 49 shy

and 0 = e Nf

- 50 shy

UK 1982 62 119 43 DOE ( 1983)

Egypt 1975 36 46 30 ILO ( 1979)

Lesotho 1973 48 88 37 ILO [ 1979)

Pakistan 1979 40 86 18 FBS [ 1983)

Sri Lanka 1981 47 97 32 DCS [19831

- 51 shy

Energy Prices

in middle income

- 52 shy

15 GJ

- 53 shy

- 54 shy

fuel can emphasize

- 55 shy

- 56 shy

OYer - 500 294 754 135 289 156 173 200 - 500 275 662 94 286 130 142 100 - 200 269 575 92 198 72 213 50 - 100 266 562 80 187 64 225 20 - 50 234 376 63 95 29 188

Under 20 244 390 67 166 1 5 143

All 266 570 93 212 85 177

~ NataraJan [19851

- 57 shy

Firewood 1979 600 409 251 17 4 121 424 1984 535 308 179 99 96 274

Soft Coke 1979 128 202 236 167 17 3 184 1984 64 180 179 152 83 153

Kerosene 1979 132 213 215 220 189 187 1984 238 369 402 382 328 357

LPG 1979 08 46 142 269 329 66 1984 152 97 83 88 101 101

Other 1979 133 131 156 170 188 139 1984 152 97 83 88 101 101

E ENERGY END-USES

- 58 shy

Cooking 826 585 503 854 797 576 833 826 489

Water heating 20 91 340 105 367 43 422

Space heating 653 324 157 91 98 57 70 131 89

- 59 shy

- 60 shy

F SUMMARY

- 61 shy

CHAPTER III

B COOKING

Consumption Ranges

- 62 shy

(par-boiling rice)

(114) Bialy 119791

225 Evans [19841

- 63 shy

- 64 shy

Charcoal stove 133 140 131 151 199

primus 69 71 10 50 45 wick 21 41 30 48 36

primus 58 57 70 57 130 wick 59 55 63 60 147

- 65 shy

Fuel Preferences

- 66 shy

- 67 shy

Stove Types

- 68 shy

- 69 shy

=c Value

Kerosene Wick

Pres sur i zed ( 0U ) 23 - 65 25 - 55 40

Gas (LPG) Butane 38 - 65 40 - 60 45

- 70 -

Stove Costs

- 71 shy

Woodburnlng Stoves

Mud 10

Clay 15 20

Metal 060 - 600

Charcoal 10 25

Kerosene 2 - 8

Gas 120

Electric 140

- 72 shy

Rubber Wood

Rice husk

Sawdust

Charcoal

Corn cob

Rice husk log

Sawdust log

Bucket

Rangsit

2-hole mud

l-ga I can

Bucket

----------------------baht-------------------shy

24 16 114 16 130

23 16 119 16 135

16 19 204 30 234

12 19 261 22 266

16 76 576 03 564

18 1 70 646 16 662

14 170 884 16 900

21 145 893 16 909

17 145 1124 16 1140

25 185 1267 16 1283

18 203 1892 16 1908

- 13 shy

D LIGHTING

- 74 shy

- 75 shy

(liters per year)

Rural

Sri Lanka

Thai land

Indonesia 1976

28 143

35 42 52 45-61 25 51

70-500 254 148

34

104 96-140

55-91

31 86

570

FBS [1983

WiJeslnghe (1984)

Natarajan (19851

Strout (19781

- 76 shy

~ lt3 3- 6 6-12

12-18 18 All

Urban lt3 3-6 6-12

12-18 18 All

25 29 41 46 51 28

29 31 31 50 86 31

087 102 144 160 179 097

103 107 107 174 302 108

90 84

104 101 106 91

45 61 48 39 39 53

156 163 205 283 322 178

164 189 243 324 425 217

088 010 088 013 110 015 137 013 153 013 096 011

075 015 089 013 104 011 130 014 167 019 096 012

~ NataraJan [19851

- 77 shy

- 78 shy

(USS)

45 43 29

large It

550 750

1175

- 19 shy

Incandescent

40 W bulb 480 60 Wbulb 850

100 W bulb 1500

Fluorescent tubes

55 W bulb 2250

120 143 149

1000 1000 1000

357 556

5000 5000

476 385

625

5000 6000 7500

41 ~7 5000

life of 20000 hours

05 05 06

130 al 130 al

130 92

250

120

- 80 shy

TOTAL

480 900 21 11

1000 5000 05 130 a

102 11 51 143

83 18 415 90

51 143 ~ 27

17 55 17 0

51 143 2075 45

2585 188

- 81 shy

Example

- 82 shy

- 83 shy

F SPACE HEATING

I shy

- 84 shy

Average Delivered

Energy for

Time c Period

B

World Bonk-31214

- 85 shy

CHAPTER IV

- 86 shy

- 87 shy

Access to Resources

- 88 shy

- 89 shy

- 90 shy

- 91 shy

Benefit Type

hydropower systems

- 92 shy

- 93 shy

- 94 shy

Supply Factors

Clear Fell ing

- 95 shy

- 96 shy

Year ($) ($) (m3) (Im3) ($) ($)

1 330 - 330 2 165 - 165 3 130 - 130 4-5 60 - 60 6 60 37 209 353 738 + 641 7-10 60 - 60

11 60 81 456 530 2417 +2276 12-15 60 - 60 16 60 73 343 706 2422 +2289 17-19 60 - 60 20 60 375 1515 882 13362 +12927

TOTALS 1645 566 2523 18939 +16728

General data

~ PFI (1981)

- 97 -

(yrs) (m3hayr)

1st year 2nd year

E citriodesra ~I

10 - 12 4 - 8

7 - 10

8 - 10 6 ms 2 - 5

5 - 15 5 - 10 5 - 10

10

4 - 5

25 - 30 15 - 20 10 - 15 8 5 - 7

7 - 10 14 - 15 8

10 15

17 - 20 10 - 15

5 - 20 35 - 60

10 - 20

25 - 60 15 odthayr 20 - 25

10 - 30 40

17 - 45 5 - 7

15 - 10

05 - 10 5 - 10 5

10

10 - 15

20 - 30 2 - 11

15

7 - 10 5 - 6

06 - 08 06 - 08

05 - 08 05 - 08

08 - 12

03 04

08 - 10 04 - 05 04 - 05 04 - 05

(lIght)

(heavy) (heavy) 05 - 060 06 - 09 06 - 08

06 06 08 - 11

07-10 07-10

Source NAS [19801

- 98 -

- 99 shy

- 100 shy

Charcoal 34 09 24 14 16

- 101 shy

Fuel

(I)

Del ivered Price

(kunit)

(2)

Net HV (MJunit)

()

(4) Effective Price

(kMJ uti I ized)

Appl lance Cost

(N=IOOk)

LPG Electricity

1471kg 251kg 3481 3481 490kg 36kWh

8-13 20-25 30-40 30-40 45-55 60-70

89 -44 -07 -02 -13 -24 -

145 58 10 27 15 27

na na 3 al

38 bl 40 45 40

(Col 2) x (Col 3)100

- 102 shy

Plantation Costs

=

1984 US$ I

0 bamboo w

- 104 shy

- 105 shy

- 106 shy

- 107 shy

E CHARCOAL

- 108 shy

Charcoal yields

- 109 shy

- 110 shy

Thin Bark

Compact Heavy

Correct Dimensions

Healthy

Low Mol sture

Source Osse (1974)

- 111 shy

Asia Thai land (1984) 009 - 021 29 03 - 07 23 13 - 30

- 112 shy

husk 16 - 45

- 113 shy

Alfalfa straw

Groundnut shells

Maize stalks

Maize cobs

Rice husks

(1 )

(1)

(2) (3)

(3)

(1) (4) ( 1 )

(4) (1)

(4) ( 1 )

(4)

(5) (4)

(5) (4) (1)

(2) (1)

(I)

(1)

( 4)

()

48

08 60

172 33

44 64 34 15 18 32 20J

192

165 149

11

85

(MJkg)

184 173 194 183 201 181 158 174 197 200 182 167 189 17 4

214 186 152 150 153 155 168 194 210 211 189 17 2

- 114 shy

- 115 shy

- 116 shy

Rice - 100 + 800 + 430

Wheat o + 600 + 290

Maize + 100 +1300 + 480

Millet o + 500 + 250

~ ICAR (1984)

- 117 shy

- 118 shy

FeedstocKs Reason

Good

Poor

high I ign in

(KWht)

Product Density

(tem3)

(tehour)

Range of Systell

Costs (US$OOOte h)

(US$ OOOte h)

30-60 NA NA

015-08 100 - 15

20shy 60 25 - 110

40 30

- 75 - 40

--

- p I I low-shaped

20 - 35 NA 20 - 60 130 - 300 30 - 60 ----

()

I

- 120 shy

- 121 shy

Residue (1) (2) (3)

Grinding

723 (422) (135 ) (150) (016)

1180 (556) (1 76) (437) (011)

10 2005 2502

1941

(338) (1403) (201)

4443

173 ( 12)

2257

1903 (1040) (411) (432) (020)

144

854 (237) (52S) (080) (012)

088 2989 3215

1941

(338) (1403) (201)

5156

150 (15)

344

1085 (239) ( 1 64) (640) (042)

144

8S4 (237) (S2S) (080) (012)

088 2171 2735

1941

(338) ( 1403) (201)

4676

174 (15)

269

- 122 shy

G ANIMAL WASTES

Direct Combustion

Animal

(kgyr)

(kg)

Cattle 27000 200 5400 80 1000 24 12

Pigs 30000 50 1500 80 300 315 18

Poultry 9000 15 13 60 5 63 63

- 124 shy

Biogas

- 125 shy

- 126 shy

CHAPTER V

B DATA SOURCES

- 127 shy

- 128 shy

cycle costs

- 129 shy

Supply Data

- 130 shy

- 131 shy

land use

As above

- 132 shy

- 133 shy

1980 1985 1990 1995 2000 2005

Standing stock 000 3 17 500 16010 13837 10827 6794 1520

Consumption 000 3yr 600 696 806 935 1084 1256

Deficit 000 3yr 250 376 529 718 948 1226

(Population ooos) (1000) ( 1 159) (1344) (1558) (1806) (2094)

Assumptions

Calculation method

- 134 shy

- 135 shy

1980 1985 1990 1995 2000 2005

open fire 75 663 504 33 196 10 stove 5 117 216 33 364 35

- 136 shy

1980 1985 1990 1995 2000 2005

- 137 shy

- 138 shy

1980 1985 1990 1995 2000 2005

500 17500

580 15907

672 14061

779 11920

903 9439

1047 6562

TOTAL 647 663 680 701 725 753

- 139 shy

Wood Consumption

~~5~ 43 2

Annual Increase 0

0 O~________L-________~________~________-L________~

1980 1985 1990 1995 2000 2005

100r--__bullbull

~~====3--- 2

1

0

o~--------~--------~----------~--------~--------~ 1980 1985 1990 1995 2000 2005

World Bank-307364

- 140 shy

- 141 shy

- 142 shy

Land type 2 3 4 5 6

Land type

Altitude (m)

200-1000 0- 500

500-1500 1000-2000

1000-2500 1500-3000

Rainfall (mm)

lt400 500-1000 500- 900

800-1200

1 000-1 300 lt1200

- 143 shy

E CASE STUDIES

Average Both fuels

548

2344 1148 3491

1482 362

1844

3826 1510 5336

- 144 shy

( of far Itany)

1145 15043 I 2850 21274 28137 44620

609 55

77 6 67

1021 119

29 34 13 14 41 27

Tota I 123069 2648 ~ 21

TOTAL 6750 3934 1560 5494 (1256)

- 145 shy

198] 1985 1987 1989 1991 993 995

Sub-Total A 00 00 530 797 13 2228 3A46 to J

- 147 shy

Annex 1

( mcwb) (MJkg)

Biomass Fuels

Fossil-Fuels

lignite Peat

Coke

Petroleum coke

40 20 15 0 5

50 13 12 9

12 12 11 24 12 40 13 12

5 5 5

10-9

155 66

200 290 82

162 160 144 152 147 154 119 164 98

179 120

31~4

293 188

113 146

201 239 218

285

352

- 148 shy

(MJkg) (t-tJ1 itre)

Fossil Fuels

Crude 01 I 086 419 367

gas

- 149 shy

(MJm3)

Refinery Gas 461

- 150 shy

Annex 2

SI American

II EnerSI Symbols

SI

AmericanGeneral

- 151 shy

W v V a A

kVA

BTUhr hp

bd bId bdoe

g kg lb lbs

tpa tpy

m km mi

2sq m mha ac

1 3cu m m

gal

SCF CF

V Biomass amp Other

od 00 odt ODT

ad AD mcwb mcdb

MAl GHV NHV

SI

AmericanGeneral

2

cubic metre 1 10000 -3 28317 -2 76455 -1 28413 -5 56826 -4 45461 -3 29574 -5 47318 -4 37854 -3 itre 99997 +2 1 28316 +1 76453 +2 28412 -2 56825 -1 45460 0 29573 -2 47316 -1 37853 0

US gallon

21134 26417

+3 +2

21134 26418

0 -1

59844 74805

+1 0

16158 20197

+3 +2

60047 75059

-2 -3

12009 15012

0 -1

960761 12009

0 0

62500 78125

-2 -3

I 12500 -1

80000 0 w

10000 22046 98421 11023

-3 0

-4 -3

10000 1

22046 98421 11023

+3

+3 -1 0

45359 45359

44643 50000

-1 -4

-4 -4

10160 10160 22400

11200

+3 0

+3

0

90718 90718 20000 89286

+2 -1 +3 -1

27778 37251

-7 -7

11630 15596

-3 -3 13410 0

74570 -1 29307 39301

-4 -4

U1 po

per minute

10000 13596

13410 14331

56869

-3 -3

-3 -2

-2

10000

13596

13410 14331

56869

+3

0

0 +1

+1

73550 73550

98632 10540

41827

+2 -1

-1 +1

+1

74570 74570 10139

1 10686

42407

+2 -1 0

+1

69780 69780 94874

93577

39683

+1 -2 -2

-2

0

17584 17584 23908

23581 25200

+1 -2 -2

-2 -1

x 10 -4

- 155 shy

Annex 4

GLOSSARY

Air-dried weight

Anaerobic processes

Bagasse

Biogas

Biomass fuels

Calorie

Coal equivalent

- 156 shy

Coal replacement

- 157 shy

- 158 shy

Utilized energy

Green weight

Annex 5

CLASS Material

Jl 0

Owner-bu i It

tinerant art isan

Industrial

I- 0shyo

marketing approach

t- 0 t-

- 162 shy

Annex 6

- 163 shy

- 164 shy

Annex 7

r I Payback Period

~-~ Time

World Bonk-307365

- 165 shy

I + F x P = f x p

- 166 shy

Annex 8

- 167 shy

- 168 shy

- 169 shy

Annex 9

for Firewood

o

are Breached

World Bank-3073612

- 171 shy

- 172 shy

- 173 shy

BIBLIOGRAPHY

- 174 -

- 115 -

- 176 -

- 177 -

- 178 shy

- 179 -

- 180 -

- 181 shy

d

- 182 -

- 183 -

sa

- 184 shy

t

- 185 shy

- 186 shy

Cover13
Abstract
Contents13

- iv shy

- v shy

TABLE or CORTEIITS

-~~~~ ~~---------------------~-------

vi

CHAPTER III

CHAPTER IV

61 61 61 61 65 65 67 67 69 70 72 73 74 74 80 82 83

85 85 86 86 87 88 88 92 92 93

93

95 97 98 98

101 102 104 107 107 109 111 112 114 117

117 120 122 122 124

- vii shy

E Case Studies 143

- viii shy

- ix shy

INTRODUCTION

- 2 shy

- 3 shy

- 4 shy

- 5 shy

- 6 shy

CHAPTER I

- 1 shy

B Primary Energy ~

C Secondary Energy

Kerosene

Production well

Refinery

(Distribution amp

Marketing)

Estimates uncertain

Energy use losses

- 8 shy

- 9 shy

Measurement Units

- 10 shy

1 bull Heat NHV

Output 1

- 11 shy

W= D(l + D100) D = W(l - W100)

- 12 shy

Fiber

Ash

Water

-

~

Net D

High E

DryG

Wet BWater

A losses

F Water

World Bank-307366

HEAT VALUE FORMULAE

- 13 shy

25

30

~ ~ 20 11

2 a

15 ic 0

15 ~ ~ ~ 8 u u i

10 J

~ ~ middot0

o 20 40 60

5

- 14 shy

- 15 shy

Heating Value

(MJkg)

20

GHV

NHV

I I MCWBo

I 30 40 60 80 100 MCDB

World Bank-307368

10 20

- 16 shy

= Vo~ume (m )

and since

then

- 11 shy

- 18 shy

Water loss 4658 kg

Water loss 13333 kg

________________________~-----~~~~------~w~a~~-r----~ ~-----66-6-7-k-9----~ 200 kg

I-

World Bank-307369

(point of use)

10 111

12000 11060 (1659)

ENERGY STAGE

Content (lIcdb)

(NHV MJkg)

Basic Data

Harvest Primary

12658 12658

526 111

12000 9620

(750)

Basic density

approx 1 800

approx 800

25 333

12000 10744 (1343)

600 kgm3

- 19 shy

-~~----------------------

- 20 shy

= Stove Cooker

030 (generation)

013

10

030

10

060

10

015 026 014

- 21 shy

Contents (E)

World Bonk-30736 10

- 22 shy

- 23 shy

- 24 shy

D BASIC STATISTICS

Data Validity

187 145 090 160 167

150 5

0367

069 197 091 068 053 141 088 085 099

8 0475

- 25 shy

Elasticities

as it of

FA b d1 d2 d3 d7

- 26 shy

- 27 -

FIGURE 16

500 -----------------------------------------------

400

300

200

100

fIIII-- fIIIIfIIII

~ -shy

--------shy-

LPG ~ ~ ~ ~

~ ~

~

o ~__________________________________________~

1974

1984

08 r-----------------------------

07 06

Kerosene Price - I

05 - - I - I shy

04

03

02 ~______________________~

1974 1984

World Bonk-31074

- 28 shy

CHAPTER II

- 29 shy

B DATA RESOURCES

- 30 shy

- 31 shy

Local Micro Surveys

- 32 shy

(kgyear)

373 232

246 161

77 87

0 3

0 0

- 33 shy

Supply variables

- 34 shy

Demand variables

o Household income

o Household size

005- 01 up to 15

05 - 1 up to 15

- 35 shy

- 36 shy

Countrysurvey

Togo LOIIe (income)

Average

35

24 33 59

30

11 2

83 236

30

51

Rllnge GJ

34 - 35

21 - 29 31 - 39 57 - 66

27 - 48

135 - 337

23 - 38

46 - 55

bull Biomass

49 - 67

26 - 72 36 - 78 70 - 84

25 - 80

79

41 lt1

22 - 87

Source

Parikh [19821

FBS (1983)

Newcombe [1980)

DeS (1983)

Grut [19711

- 37 shy

- 38 shy

(1 vi II age) 22 94 - 38 Spears [1978)

- 39 shy

- 40 shy

- 41 shy

bullbull bull

bull bull bull

- 42 shy

Rural so

Srazil

~

J bull bull

bull I

I bull

bull

~ I (

I

OL--L~__L--L~__~~~__~~~__~~~__~~~~

bull

bull _ Pakistan

India

bull -- - - ~r- _~ ~ ~ ------------------------------~B~ra~zil

bullbull

- 43 shy

Rurol 100

Indio

Pakistan CD

805s () gtshy ~ w

QZ J

~ in

o 2 4 6 8 10 12 14 16 18

Urban

80

I India

20

o 2 4 6 8 10 12 14 16 18

World Bonk-307362

- 44 shy

rural areas but

- 45 shy

biofuels -

other

to saturate at the

increases shortages

Household Size

- 46 shy

-----

---------

- 47 shy

f

Total

1

_ Per Capita

Household Size --

f ~ ltJ)c w

Total

Per Capita

Household Size ---t

World Bank-307363

bull bullbull

- 48 shy

Rural

bull 10 -

Sri Lanka bull8

( Q)

~ (] gt 6 Indio

~ c bull

1bull~ -_ shyw _-shy __ ~ 0 0 4 U (j) 0

2

o 2 4 6 8 10 12 14 16 18

Urban

gt 60gt ee

(j) c w

--r ----shy~ ---__-_ - 2

O~~~__~~~__L-~~__L-~~__L-~~__L-~~~

o 2 4 6 8 10 12 14 16 18

World Bank-3073611

- 49 shy

and 0 = e Nf

- 50 shy

UK 1982 62 119 43 DOE ( 1983)

Egypt 1975 36 46 30 ILO ( 1979)

Lesotho 1973 48 88 37 ILO [ 1979)

Pakistan 1979 40 86 18 FBS [ 1983)

Sri Lanka 1981 47 97 32 DCS [19831

- 51 shy

Energy Prices

in middle income

- 52 shy

15 GJ

- 53 shy

- 54 shy

fuel can emphasize

- 55 shy

- 56 shy

OYer - 500 294 754 135 289 156 173 200 - 500 275 662 94 286 130 142 100 - 200 269 575 92 198 72 213 50 - 100 266 562 80 187 64 225 20 - 50 234 376 63 95 29 188

Under 20 244 390 67 166 1 5 143

All 266 570 93 212 85 177

~ NataraJan [19851

- 57 shy

Firewood 1979 600 409 251 17 4 121 424 1984 535 308 179 99 96 274

Soft Coke 1979 128 202 236 167 17 3 184 1984 64 180 179 152 83 153

Kerosene 1979 132 213 215 220 189 187 1984 238 369 402 382 328 357

LPG 1979 08 46 142 269 329 66 1984 152 97 83 88 101 101

Other 1979 133 131 156 170 188 139 1984 152 97 83 88 101 101

E ENERGY END-USES

- 58 shy

Cooking 826 585 503 854 797 576 833 826 489

Water heating 20 91 340 105 367 43 422

Space heating 653 324 157 91 98 57 70 131 89

- 59 shy

- 60 shy

F SUMMARY

- 61 shy

CHAPTER III

B COOKING

Consumption Ranges

- 62 shy

(par-boiling rice)

(114) Bialy 119791

225 Evans [19841

- 63 shy

- 64 shy

Charcoal stove 133 140 131 151 199

primus 69 71 10 50 45 wick 21 41 30 48 36

primus 58 57 70 57 130 wick 59 55 63 60 147

- 65 shy

Fuel Preferences

- 66 shy

- 67 shy

Stove Types

- 68 shy

- 69 shy

=c Value

Kerosene Wick

Pres sur i zed ( 0U ) 23 - 65 25 - 55 40

Gas (LPG) Butane 38 - 65 40 - 60 45

- 70 -

Stove Costs

- 71 shy

Woodburnlng Stoves

Mud 10

Clay 15 20

Metal 060 - 600

Charcoal 10 25

Kerosene 2 - 8

Gas 120

Electric 140

- 72 shy

Rubber Wood

Rice husk

Sawdust

Charcoal

Corn cob

Rice husk log

Sawdust log

Bucket

Rangsit

2-hole mud

l-ga I can

Bucket

----------------------baht-------------------shy

24 16 114 16 130

23 16 119 16 135

16 19 204 30 234

12 19 261 22 266

16 76 576 03 564

18 1 70 646 16 662

14 170 884 16 900

21 145 893 16 909

17 145 1124 16 1140

25 185 1267 16 1283

18 203 1892 16 1908

- 13 shy

D LIGHTING

- 74 shy

- 75 shy

(liters per year)

Rural

Sri Lanka

Thai land

Indonesia 1976

28 143

35 42 52 45-61 25 51

70-500 254 148

34

104 96-140

55-91

31 86

570

FBS [1983

WiJeslnghe (1984)

Natarajan (19851

Strout (19781

- 76 shy

~ lt3 3- 6 6-12

12-18 18 All

Urban lt3 3-6 6-12

12-18 18 All

25 29 41 46 51 28

29 31 31 50 86 31

087 102 144 160 179 097

103 107 107 174 302 108

90 84

104 101 106 91

45 61 48 39 39 53

156 163 205 283 322 178

164 189 243 324 425 217

088 010 088 013 110 015 137 013 153 013 096 011

075 015 089 013 104 011 130 014 167 019 096 012

~ NataraJan [19851

- 77 shy

- 78 shy

(USS)

45 43 29

large It

550 750

1175

- 19 shy

Incandescent

40 W bulb 480 60 Wbulb 850

100 W bulb 1500

Fluorescent tubes

55 W bulb 2250

120 143 149

1000 1000 1000

357 556

5000 5000

476 385

625

5000 6000 7500

41 ~7 5000

life of 20000 hours

05 05 06

130 al 130 al

130 92

250

120

- 80 shy

TOTAL

480 900 21 11

1000 5000 05 130 a

102 11 51 143

83 18 415 90

51 143 ~ 27

17 55 17 0

51 143 2075 45

2585 188

- 81 shy

Example

- 82 shy

- 83 shy

F SPACE HEATING

I shy

- 84 shy

Average Delivered

Energy for

Time c Period

B

World Bonk-31214

- 85 shy

CHAPTER IV

- 86 shy

- 87 shy

Access to Resources

- 88 shy

- 89 shy

- 90 shy

- 91 shy

Benefit Type

hydropower systems

- 92 shy

- 93 shy

- 94 shy

Supply Factors

Clear Fell ing

- 95 shy

- 96 shy

Year ($) ($) (m3) (Im3) ($) ($)

1 330 - 330 2 165 - 165 3 130 - 130 4-5 60 - 60 6 60 37 209 353 738 + 641 7-10 60 - 60

11 60 81 456 530 2417 +2276 12-15 60 - 60 16 60 73 343 706 2422 +2289 17-19 60 - 60 20 60 375 1515 882 13362 +12927

TOTALS 1645 566 2523 18939 +16728

General data

~ PFI (1981)

- 97 -

(yrs) (m3hayr)

1st year 2nd year

E citriodesra ~I

10 - 12 4 - 8

7 - 10

8 - 10 6 ms 2 - 5

5 - 15 5 - 10 5 - 10

10

4 - 5

25 - 30 15 - 20 10 - 15 8 5 - 7

7 - 10 14 - 15 8

10 15

17 - 20 10 - 15

5 - 20 35 - 60

10 - 20

25 - 60 15 odthayr 20 - 25

10 - 30 40

17 - 45 5 - 7

15 - 10

05 - 10 5 - 10 5

10

10 - 15

20 - 30 2 - 11

15

7 - 10 5 - 6

06 - 08 06 - 08

05 - 08 05 - 08

08 - 12

03 04

08 - 10 04 - 05 04 - 05 04 - 05

(lIght)

(heavy) (heavy) 05 - 060 06 - 09 06 - 08

06 06 08 - 11

07-10 07-10

Source NAS [19801

- 98 -

- 99 shy

- 100 shy

Charcoal 34 09 24 14 16

- 101 shy

Fuel

(I)

Del ivered Price

(kunit)

(2)

Net HV (MJunit)

()

(4) Effective Price

(kMJ uti I ized)

Appl lance Cost

(N=IOOk)

LPG Electricity

1471kg 251kg 3481 3481 490kg 36kWh

8-13 20-25 30-40 30-40 45-55 60-70

89 -44 -07 -02 -13 -24 -

145 58 10 27 15 27

na na 3 al

38 bl 40 45 40

(Col 2) x (Col 3)100

- 102 shy

Plantation Costs

=

1984 US$ I

0 bamboo w

- 104 shy

- 105 shy

- 106 shy

- 107 shy

E CHARCOAL

- 108 shy

Charcoal yields

- 109 shy

- 110 shy

Thin Bark

Compact Heavy

Correct Dimensions

Healthy

Low Mol sture

Source Osse (1974)

- 111 shy

Asia Thai land (1984) 009 - 021 29 03 - 07 23 13 - 30

- 112 shy

husk 16 - 45

- 113 shy

Alfalfa straw

Groundnut shells

Maize stalks

Maize cobs

Rice husks

(1 )

(1)

(2) (3)

(3)

(1) (4) ( 1 )

(4) (1)

(4) ( 1 )

(4)

(5) (4)

(5) (4) (1)

(2) (1)

(I)

(1)

( 4)

()

48

08 60

172 33

44 64 34 15 18 32 20J

192

165 149

11

85

(MJkg)

184 173 194 183 201 181 158 174 197 200 182 167 189 17 4

214 186 152 150 153 155 168 194 210 211 189 17 2

- 114 shy

- 115 shy

- 116 shy

Rice - 100 + 800 + 430

Wheat o + 600 + 290

Maize + 100 +1300 + 480

Millet o + 500 + 250

~ ICAR (1984)

- 117 shy

- 118 shy

FeedstocKs Reason

Good

Poor

high I ign in

(KWht)

Product Density

(tem3)

(tehour)

Range of Systell

Costs (US$OOOte h)

(US$ OOOte h)

30-60 NA NA

015-08 100 - 15

20shy 60 25 - 110

40 30

- 75 - 40

--

- p I I low-shaped

20 - 35 NA 20 - 60 130 - 300 30 - 60 ----

()

I

- 120 shy

- 121 shy

Residue (1) (2) (3)

Grinding

723 (422) (135 ) (150) (016)

1180 (556) (1 76) (437) (011)

10 2005 2502

1941

(338) (1403) (201)

4443

173 ( 12)

2257

1903 (1040) (411) (432) (020)

144

854 (237) (52S) (080) (012)

088 2989 3215

1941

(338) (1403) (201)

5156

150 (15)

344

1085 (239) ( 1 64) (640) (042)

144

8S4 (237) (S2S) (080) (012)

088 2171 2735

1941

(338) ( 1403) (201)

4676

174 (15)

269

- 122 shy

G ANIMAL WASTES

Direct Combustion

Animal

(kgyr)

(kg)

Cattle 27000 200 5400 80 1000 24 12

Pigs 30000 50 1500 80 300 315 18

Poultry 9000 15 13 60 5 63 63

- 124 shy

Biogas

- 125 shy

- 126 shy

CHAPTER V

B DATA SOURCES

- 127 shy

- 128 shy

cycle costs

- 129 shy

Supply Data

- 130 shy

- 131 shy

land use

As above

- 132 shy

- 133 shy

1980 1985 1990 1995 2000 2005

Standing stock 000 3 17 500 16010 13837 10827 6794 1520

Consumption 000 3yr 600 696 806 935 1084 1256

Deficit 000 3yr 250 376 529 718 948 1226

(Population ooos) (1000) ( 1 159) (1344) (1558) (1806) (2094)

Assumptions

Calculation method

- 134 shy

- 135 shy

1980 1985 1990 1995 2000 2005

open fire 75 663 504 33 196 10 stove 5 117 216 33 364 35

- 136 shy

1980 1985 1990 1995 2000 2005

- 137 shy

- 138 shy

1980 1985 1990 1995 2000 2005

500 17500

580 15907

672 14061

779 11920

903 9439

1047 6562

TOTAL 647 663 680 701 725 753

- 139 shy

Wood Consumption

~~5~ 43 2

Annual Increase 0

0 O~________L-________~________~________-L________~

1980 1985 1990 1995 2000 2005

100r--__bullbull

~~====3--- 2

1

0

o~--------~--------~----------~--------~--------~ 1980 1985 1990 1995 2000 2005

World Bank-307364

- 140 shy

- 141 shy

- 142 shy

Land type 2 3 4 5 6

Land type

Altitude (m)

200-1000 0- 500

500-1500 1000-2000

1000-2500 1500-3000

Rainfall (mm)

lt400 500-1000 500- 900

800-1200

1 000-1 300 lt1200

- 143 shy

E CASE STUDIES

Average Both fuels

548

2344 1148 3491

1482 362

1844

3826 1510 5336

- 144 shy

( of far Itany)

1145 15043 I 2850 21274 28137 44620

609 55

77 6 67

1021 119

29 34 13 14 41 27

Tota I 123069 2648 ~ 21

TOTAL 6750 3934 1560 5494 (1256)

- 145 shy

198] 1985 1987 1989 1991 993 995

Sub-Total A 00 00 530 797 13 2228 3A46 to J

- 147 shy

Annex 1

( mcwb) (MJkg)

Biomass Fuels

Fossil-Fuels

lignite Peat

Coke

Petroleum coke

40 20 15 0 5

50 13 12 9

12 12 11 24 12 40 13 12

5 5 5

10-9

155 66

200 290 82

162 160 144 152 147 154 119 164 98

179 120

31~4

293 188

113 146

201 239 218

285

352

- 148 shy

(MJkg) (t-tJ1 itre)

Fossil Fuels

Crude 01 I 086 419 367

gas

- 149 shy

(MJm3)

Refinery Gas 461

- 150 shy

Annex 2

SI American

II EnerSI Symbols

SI

AmericanGeneral

- 151 shy

W v V a A

kVA

BTUhr hp

bd bId bdoe

g kg lb lbs

tpa tpy

m km mi

2sq m mha ac

1 3cu m m

gal

SCF CF

V Biomass amp Other

od 00 odt ODT

ad AD mcwb mcdb

MAl GHV NHV

SI

AmericanGeneral

2

cubic metre 1 10000 -3 28317 -2 76455 -1 28413 -5 56826 -4 45461 -3 29574 -5 47318 -4 37854 -3 itre 99997 +2 1 28316 +1 76453 +2 28412 -2 56825 -1 45460 0 29573 -2 47316 -1 37853 0

US gallon

21134 26417

+3 +2

21134 26418

0 -1

59844 74805

+1 0

16158 20197

+3 +2

60047 75059

-2 -3

12009 15012